{ "mac_mini": { "warm_results": [ { "request_id": "warm-mac_mini-1", "model": "qwen2.5:7b", "ok": true, "started_at": 1778137876.7394822, "latency_s": 1.569229250017088, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "warm-mac_mini-2", "model": "qwen3:30b", "ok": true, "started_at": 1778137878.3087842, "latency_s": 12.374412957986351, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "warm-mac_mini-3", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778137890.6834922, "latency_s": 17.03599041700363, "finish_reason": "length", "content_preview": "- **Batch & pipeline requests** \u2013 group incoming messages and process them in parallel pipelines to amortize GPU kernel ", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 } ], "steps": [ { "summary": { "concurrency": 1, "total_requests": 9, "request_count": 9, "success_count": 9, "error_count": 0, "error_rate": 0.0, "wall_s": 87.45133516698843, "throughput_rps": 0.10291438069887085, "latency_avg_s": 9.716113037099907, "latency_p50_s": 10.071201917016879, "latency_p95_s": 13.570814591576346, "latency_max_s": 14.102356791961938, "completion_tps": 7.60422924052768, "sample_errors": [], "peak_cpu_busy_pct": 67.17999999999999, "peak_load1": 7.07, "min_memory_free_pct": 15.0, "thermal_warning_seen": false, "performance_warning_seen": false, "thermal_status_peak": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded" }, "requests": [ { "request_id": "mac_mini-1-1", "model": "qwen2.5:7b", "ok": true, "started_at": 1778137907.9312792, "latency_s": 5.249839291966055, "finish_reason": "stop", "content_preview": "- Optimize model efficiency: Reduce latency with model quantization or pruning.\n- Implement load balancing: Distribute i", "prompt_tokens": 77, "completion_tokens": 41, "total_tokens": 118 }, { "request_id": "mac_mini-1-2", "model": "qwen3:30b", "ok": true, "started_at": 1778137913.181327, "latency_s": 9.296648624993395, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster and wants three short operational tip", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-1-3", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778137922.478168, "latency_s": 12.773501290997956, "finish_reason": "length", "content_preview": "- **Batch requests & use async I/O** \u2013 Process multiple user messages in parallel pipelines to avoid blocking a single t", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-1-4", "model": "qwen2.5:7b", "ok": true, "started_at": 1778137935.251866, "latency_s": 5.382027915969957, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-1-5", "model": "qwen3:30b", "ok": true, "started_at": 1778137940.634131, "latency_s": 10.071201917016879, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-1-6", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778137950.705482, "latency_s": 14.102356791961938, "finish_reason": "length", "content_preview": "- **Scale horizontally with auto\u2011scale groups** \u2013 add instances when request rate spikes, and keep them warm to avoid co", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-1-7", "model": "qwen2.5:7b", "ok": true, "started_at": 1778137964.8080168, "latency_s": 10.131589416996576, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-1-8", "model": "qwen3:30b", "ok": true, "started_at": 1778137974.939845, "latency_s": 8.941197249980178, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-1-9", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778137983.8811882, "latency_s": 11.496654834016226, "finish_reason": "length", "content_preview": "- **", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 } ], "samples": [ { "timestamp": 1778137907.7199202, "unit": "mac_mini", "load1": 5.83, "load5": 5.79, "load15": 4.81, "cpu_user_pct": 21.45, "cpu_sys_pct": 12.87, "cpu_idle_pct": 65.66, "cpu_busy_pct": 34.32, "phys_mem_used": "47G", "phys_mem_unused": "511M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137910.970124, "unit": "mac_mini", "load1": 5.68, "load5": 5.76, "load15": 4.81, "cpu_user_pct": 22.97, "cpu_sys_pct": 17.44, "cpu_idle_pct": 59.57, "cpu_busy_pct": 40.41, "phys_mem_used": "47G", "phys_mem_unused": "283M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137914.271209, "unit": "mac_mini", "load1": 5.68, "load5": 5.76, "load15": 4.81, "cpu_user_pct": 21.79, "cpu_sys_pct": 15.81, "cpu_idle_pct": 62.39, "cpu_busy_pct": 37.6, "phys_mem_used": "41G", "phys_mem_unused": "6529M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137917.521121, "unit": "mac_mini", "load1": 5.47, "load5": 5.72, "load15": 4.8, "cpu_user_pct": 36.75, "cpu_sys_pct": 22.22, "cpu_idle_pct": 41.2, "cpu_busy_pct": 58.97, "phys_mem_used": "44G", "phys_mem_unused": "3340M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137920.772452, "unit": "mac_mini", "load1": 5.59, "load5": 5.74, "load15": 4.81, "cpu_user_pct": 26.49, "cpu_sys_pct": 19.23, "cpu_idle_pct": 54.27, "cpu_busy_pct": 45.72, "phys_mem_used": "46G", "phys_mem_unused": "1502M", "memory_free_pct": 36.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137924.030206, "unit": "mac_mini", "load1": 5.59, "load5": 5.74, "load15": 4.81, "cpu_user_pct": 19.91, "cpu_sys_pct": 17.84, "cpu_idle_pct": 62.24, "cpu_busy_pct": 37.75, "phys_mem_used": "47G", "phys_mem_unused": "595M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137927.293612, "unit": "mac_mini", "load1": 5.54, "load5": 5.73, "load15": 4.81, "cpu_user_pct": 22.4, "cpu_sys_pct": 18.67, "cpu_idle_pct": 58.92, "cpu_busy_pct": 41.07, "phys_mem_used": "47G", "phys_mem_unused": "609M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137930.5966232, "unit": "mac_mini", "load1": 5.5, "load5": 5.71, "load15": 4.81, "cpu_user_pct": 22.78, "cpu_sys_pct": 18.14, "cpu_idle_pct": 59.7, "cpu_busy_pct": 40.92, "phys_mem_used": "47G", "phys_mem_unused": "645M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137933.860318, "unit": "mac_mini", "load1": 5.5, "load5": 5.71, "load15": 4.81, "cpu_user_pct": 15.25, "cpu_sys_pct": 11.86, "cpu_idle_pct": 72.88, "cpu_busy_pct": 27.11, "phys_mem_used": "47G", "phys_mem_unused": "647M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137937.116744, "unit": "mac_mini", "load1": 5.14, "load5": 5.64, "load15": 4.79, "cpu_user_pct": 13.98, "cpu_sys_pct": 20.33, "cpu_idle_pct": 65.67, "cpu_busy_pct": 34.31, "phys_mem_used": "47G", "phys_mem_unused": "425M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137940.395255, "unit": "mac_mini", "load1": 5.14, "load5": 5.64, "load15": 4.79, "cpu_user_pct": 17.91, "cpu_sys_pct": 14.58, "cpu_idle_pct": 67.5, "cpu_busy_pct": 32.49, "phys_mem_used": "47G", "phys_mem_unused": "378M", "memory_free_pct": 15.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137943.657977, "unit": "mac_mini", "load1": 5.13, "load5": 5.62, "load15": 4.79, "cpu_user_pct": 12.29, "cpu_sys_pct": 15.98, "cpu_idle_pct": 71.72, "cpu_busy_pct": 28.27, "phys_mem_used": "43G", "phys_mem_unused": "4453M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137946.9358552, "unit": "mac_mini", "load1": 4.8, "load5": 5.55, "load15": 4.77, "cpu_user_pct": 40.62, "cpu_sys_pct": 26.56, "cpu_idle_pct": 32.81, "cpu_busy_pct": 67.17999999999999, "phys_mem_used": "46G", "phys_mem_unused": "1135M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137950.221809, "unit": "mac_mini", "load1": 4.8, "load5": 5.55, "load15": 4.77, "cpu_user_pct": 31.55, "cpu_sys_pct": 15.16, "cpu_idle_pct": 53.27, "cpu_busy_pct": 46.71, "phys_mem_used": "47G", "phys_mem_unused": "890M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137953.485321, "unit": "mac_mini", "load1": 5.05, "load5": 5.59, "load15": 4.79, "cpu_user_pct": 16.0, "cpu_sys_pct": 18.0, "cpu_idle_pct": 66.0, "cpu_busy_pct": 34.0, "phys_mem_used": "47G", "phys_mem_unused": "607M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137956.7459872, "unit": "mac_mini", "load1": 5.05, "load5": 5.58, "load15": 4.79, "cpu_user_pct": 19.91, "cpu_sys_pct": 17.42, "cpu_idle_pct": 62.65, "cpu_busy_pct": 37.33, "phys_mem_used": "47G", "phys_mem_unused": "627M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137960.01851, "unit": "mac_mini", "load1": 5.05, "load5": 5.58, "load15": 4.79, "cpu_user_pct": 20.81, "cpu_sys_pct": 15.91, "cpu_idle_pct": 63.26, "cpu_busy_pct": 36.72, "phys_mem_used": "47G", "phys_mem_unused": "682M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137963.2841558, "unit": "mac_mini", "load1": 5.04, "load5": 5.57, "load15": 4.79, "cpu_user_pct": 23.93, "cpu_sys_pct": 14.52, "cpu_idle_pct": 61.53, "cpu_busy_pct": 38.45, "phys_mem_used": "47G", "phys_mem_unused": "551M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137966.539916, "unit": "mac_mini", "load1": 5.6, "load5": 5.67, "load15": 4.83, "cpu_user_pct": 17.62, "cpu_sys_pct": 23.77, "cpu_idle_pct": 58.6, "cpu_busy_pct": 41.39, "phys_mem_used": "47G", "phys_mem_unused": "335M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137969.891623, "unit": "mac_mini", "load1": 5.6, "load5": 5.67, "load15": 4.83, "cpu_user_pct": 23.36, "cpu_sys_pct": 24.59, "cpu_idle_pct": 52.4, "cpu_busy_pct": 47.95, "phys_mem_used": "47G", "phys_mem_unused": "519M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137973.149695, "unit": "mac_mini", "load1": 5.79, "load5": 5.71, "load15": 4.85, "cpu_user_pct": 25.1, "cpu_sys_pct": 20.85, "cpu_idle_pct": 54.4, "cpu_busy_pct": 45.95, "phys_mem_used": "47G", "phys_mem_unused": "357M", "memory_free_pct": 23.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137976.3976161, "unit": "mac_mini", "load1": 6.29, "load5": 5.82, "load15": 4.89, "cpu_user_pct": 27.8, "cpu_sys_pct": 21.25, "cpu_idle_pct": 51.66, "cpu_busy_pct": 49.05, "phys_mem_used": "41G", "phys_mem_unused": "6497M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137979.655377, "unit": "mac_mini", "load1": 6.29, "load5": 5.82, "load15": 4.89, "cpu_user_pct": 21.55, "cpu_sys_pct": 18.53, "cpu_idle_pct": 59.91, "cpu_busy_pct": 40.08, "phys_mem_used": "44G", "phys_mem_unused": "3498M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137982.943807, "unit": "mac_mini", "load1": 7.07, "load5": 5.99, "load15": 4.96, "cpu_user_pct": 36.75, "cpu_sys_pct": 17.9, "cpu_idle_pct": 46.15, "cpu_busy_pct": 54.65, "phys_mem_used": "46G", "phys_mem_unused": "1243M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137986.194933, "unit": "mac_mini", "load1": 6.74, "load5": 5.94, "load15": 4.94, "cpu_user_pct": 19.4, "cpu_sys_pct": 18.14, "cpu_idle_pct": 62.44, "cpu_busy_pct": 37.54, "phys_mem_used": "47G", "phys_mem_unused": "674M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137989.4540539, "unit": "mac_mini", "load1": 6.74, "load5": 5.94, "load15": 4.94, "cpu_user_pct": 20.25, "cpu_sys_pct": 18.14, "cpu_idle_pct": 61.6, "cpu_busy_pct": 38.39, "phys_mem_used": "47G", "phys_mem_unused": "648M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137992.713684, "unit": "mac_mini", "load1": 6.36, "load5": 5.87, "load15": 4.93, "cpu_user_pct": 21.0, "cpu_sys_pct": 13.86, "cpu_idle_pct": 65.12, "cpu_busy_pct": 34.86, "phys_mem_used": "47G", "phys_mem_unused": "595M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778137995.583227, "unit": "mac_mini", "load1": 6.01, "load5": 5.81, "load15": 4.91, "cpu_user_pct": 24.3, "cpu_sys_pct": 15.87, "cpu_idle_pct": 60.8, "cpu_busy_pct": 40.17, "phys_mem_used": "47G", "phys_mem_unused": "399M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 } ] }, { "summary": { "concurrency": 2, "total_requests": 9, "request_count": 9, "success_count": 9, "error_count": 0, "error_rate": 0.0, "wall_s": 73.10480725002708, "throughput_rps": 0.12311091894708014, "latency_avg_s": 14.896224333322607, "latency_p50_s": 13.752180750016123, "latency_p95_s": 21.623434282978998, "latency_max_s": 21.790378832956776, "completion_tps": 8.97341809214273, "sample_errors": [], "peak_cpu_busy_pct": 64.6, "peak_load1": 6.68, "min_memory_free_pct": 13.0, "thermal_warning_seen": false, "performance_warning_seen": false, "thermal_status_peak": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded" }, "requests": [ { "request_id": "mac_mini-2-1", "model": "qwen2.5:7b", "ok": true, "started_at": 1778137998.034049, "latency_s": 4.703316375031136, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-2-2", "model": "qwen3:30b", "ok": true, "started_at": 1778137998.0342631, "latency_s": 13.336477916978765, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They need three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-2-3", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138002.737546, "latency_s": 21.082235333975405, "finish_reason": "length", "content_preview": "- **Batch requests & use async I/O** \u2013 Process multiple incoming messages in parallel and return responses as soon as ea", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-2-4", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138011.370941, "latency_s": 13.752180750016123, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-2-5", "model": "qwen3:30b", "ok": true, "started_at": 1778138023.8200588, "latency_s": 10.15448537498014, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-2-6", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138025.123323, "latency_s": 21.790378832956776, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages and process them in parallel batches to maximize GPU utilization. \n- **Li", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-2-7", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138033.97474, "latency_s": 15.790515417000279, "finish_reason": "stop", "content_preview": "- Optimize model efficiency: Reduce inference latency.\n- Implement load balancing: Distribute traffic evenly.\n- Use cach", "prompt_tokens": 77, "completion_tokens": 32, "total_tokens": 109 }, { "request_id": "mac_mini-2-8", "model": "qwen3:30b", "ok": true, "started_at": 1778138046.9139168, "latency_s": 12.083411540952511, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-2-9", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138049.765472, "latency_s": 21.373017458012328, "finish_reason": "length", "content_preview": "- **Batch requests & use async I/O** \u2013 Process multiple incoming messages in parallel and avoid blocking calls. \n- **Sc", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 } ], "samples": [ { "timestamp": 1778137997.8285959, "unit": "mac_mini", "load1": 6.01, "load5": 5.81, "load15": 4.91, "cpu_user_pct": 19.6, "cpu_sys_pct": 11.86, "cpu_idle_pct": 69.6, "cpu_busy_pct": 31.46, "phys_mem_used": "47G", "phys_mem_unused": "339M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138001.0758998, "unit": "mac_mini", "load1": 5.85, "load5": 5.78, "load15": 4.9, "cpu_user_pct": 16.86, "cpu_sys_pct": 16.47, "cpu_idle_pct": 66.66, "cpu_busy_pct": 33.33, "phys_mem_used": "47G", "phys_mem_unused": "198M", "memory_free_pct": 17.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138004.3436449, "unit": "mac_mini", "load1": 5.85, "load5": 5.78, "load15": 4.9, "cpu_user_pct": 17.37, "cpu_sys_pct": 19.49, "cpu_idle_pct": 63.13, "cpu_busy_pct": 36.86, "phys_mem_used": "44G", "phys_mem_unused": "3675M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138007.598489, "unit": "mac_mini", "load1": 6.02, "load5": 5.81, "load15": 4.92, "cpu_user_pct": 20.76, "cpu_sys_pct": 19.6, "cpu_idle_pct": 60.16, "cpu_busy_pct": 40.36, "phys_mem_used": "46G", "phys_mem_unused": "1647M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138010.859754, "unit": "mac_mini", "load1": 5.94, "load5": 5.8, "load15": 4.92, "cpu_user_pct": 31.38, "cpu_sys_pct": 18.82, "cpu_idle_pct": 49.79, "cpu_busy_pct": 50.2, "phys_mem_used": "46G", "phys_mem_unused": "1538M", "memory_free_pct": 36.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138014.117068, "unit": "mac_mini", "load1": 5.94, "load5": 5.8, "load15": 4.92, "cpu_user_pct": 15.98, "cpu_sys_pct": 18.3, "cpu_idle_pct": 65.98, "cpu_busy_pct": 34.28, "phys_mem_used": "47G", "phys_mem_unused": "611M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138017.381909, "unit": "mac_mini", "load1": 5.7, "load5": 5.75, "load15": 4.91, "cpu_user_pct": 24.58, "cpu_sys_pct": 17.5, "cpu_idle_pct": 57.91, "cpu_busy_pct": 42.08, "phys_mem_used": "47G", "phys_mem_unused": "632M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138020.6441731, "unit": "mac_mini", "load1": 5.65, "load5": 5.74, "load15": 4.91, "cpu_user_pct": 17.8, "cpu_sys_pct": 16.66, "cpu_idle_pct": 66.25, "cpu_busy_pct": 34.46, "phys_mem_used": "47G", "phys_mem_unused": "645M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138023.906967, "unit": "mac_mini", "load1": 5.65, "load5": 5.74, "load15": 4.91, "cpu_user_pct": 21.7, "cpu_sys_pct": 20.0, "cpu_idle_pct": 58.29, "cpu_busy_pct": 41.7, "phys_mem_used": "47G", "phys_mem_unused": "297M", "memory_free_pct": 13.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138027.163885, "unit": "mac_mini", "load1": 5.52, "load5": 5.71, "load15": 4.91, "cpu_user_pct": 19.24, "cpu_sys_pct": 18.82, "cpu_idle_pct": 61.92, "cpu_busy_pct": 38.06, "phys_mem_used": "44G", "phys_mem_unused": "3502M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138030.429329, "unit": "mac_mini", "load1": 5.52, "load5": 5.71, "load15": 4.91, "cpu_user_pct": 24.89, "cpu_sys_pct": 25.72, "cpu_idle_pct": 49.37, "cpu_busy_pct": 50.61, "phys_mem_used": "46G", "phys_mem_unused": "1051M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138033.701418, "unit": "mac_mini", "load1": 6.68, "load5": 5.95, "load15": 4.99, "cpu_user_pct": 27.7, "cpu_sys_pct": 18.77, "cpu_idle_pct": 54.14, "cpu_busy_pct": 46.47, "phys_mem_used": "46G", "phys_mem_unused": "1525M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138036.951359, "unit": "mac_mini", "load1": 6.3, "load5": 5.88, "load15": 4.98, "cpu_user_pct": 20.83, "cpu_sys_pct": 21.66, "cpu_idle_pct": 57.5, "cpu_busy_pct": 42.489999999999995, "phys_mem_used": "47G", "phys_mem_unused": "569M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138040.210894, "unit": "mac_mini", "load1": 6.3, "load5": 5.88, "load15": 4.98, "cpu_user_pct": 15.31, "cpu_sys_pct": 21.27, "cpu_idle_pct": 63.4, "cpu_busy_pct": 36.58, "phys_mem_used": "47G", "phys_mem_unused": "654M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138043.468818, "unit": "mac_mini", "load1": 6.12, "load5": 5.85, "load15": 4.97, "cpu_user_pct": 35.84, "cpu_sys_pct": 20.37, "cpu_idle_pct": 43.77, "cpu_busy_pct": 56.21000000000001, "phys_mem_used": "47G", "phys_mem_unused": "499M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138046.7516859, "unit": "mac_mini", "load1": 6.35, "load5": 5.9, "load15": 4.99, "cpu_user_pct": 24.47, "cpu_sys_pct": 19.83, "cpu_idle_pct": 55.69, "cpu_busy_pct": 44.3, "phys_mem_used": "47G", "phys_mem_unused": "614M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138050.0123498, "unit": "mac_mini", "load1": 6.35, "load5": 5.9, "load15": 4.99, "cpu_user_pct": 41.5, "cpu_sys_pct": 23.1, "cpu_idle_pct": 35.47, "cpu_busy_pct": 64.6, "phys_mem_used": "43G", "phys_mem_unused": "4213M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138053.352143, "unit": "mac_mini", "load1": 6.48, "load5": 5.94, "load15": 5.01, "cpu_user_pct": 19.75, "cpu_sys_pct": 17.28, "cpu_idle_pct": 62.96, "cpu_busy_pct": 37.03, "phys_mem_used": "45G", "phys_mem_unused": "2570M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138056.611943, "unit": "mac_mini", "load1": 6.28, "load5": 5.9, "load15": 5.01, "cpu_user_pct": 18.58, "cpu_sys_pct": 16.37, "cpu_idle_pct": 65.4, "cpu_busy_pct": 34.95, "phys_mem_used": "47G", "phys_mem_unused": "599M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138059.867125, "unit": "mac_mini", "load1": 6.28, "load5": 5.9, "load15": 5.01, "cpu_user_pct": 16.94, "cpu_sys_pct": 19.83, "cpu_idle_pct": 63.22, "cpu_busy_pct": 36.769999999999996, "phys_mem_used": "47G", "phys_mem_unused": "659M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138063.13375, "unit": "mac_mini", "load1": 6.02, "load5": 5.86, "load15": 4.99, "cpu_user_pct": 17.64, "cpu_sys_pct": 16.8, "cpu_idle_pct": 65.54, "cpu_busy_pct": 34.44, "phys_mem_used": "47G", "phys_mem_unused": "531M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138066.3920882, "unit": "mac_mini", "load1": 6.02, "load5": 5.86, "load15": 5.0, "cpu_user_pct": 17.28, "cpu_sys_pct": 18.1, "cpu_idle_pct": 64.6, "cpu_busy_pct": 35.38, "phys_mem_used": "47G", "phys_mem_unused": "610M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138069.653284, "unit": "mac_mini", "load1": 6.02, "load5": 5.86, "load15": 5.0, "cpu_user_pct": 24.59, "cpu_sys_pct": 15.57, "cpu_idle_pct": 59.83, "cpu_busy_pct": 40.16, "phys_mem_used": "47G", "phys_mem_unused": "572M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138071.343934, "unit": "mac_mini", "load1": 5.69, "load5": 5.79, "load15": 4.98, "cpu_user_pct": 28.63, "cpu_sys_pct": 14.52, "cpu_idle_pct": 56.84, "cpu_busy_pct": 43.15, "phys_mem_used": "47G", "phys_mem_unused": "613M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 } ] }, { "summary": { "concurrency": 3, "total_requests": 12, "request_count": 12, "success_count": 12, "error_count": 0, "error_rate": 0.0, "wall_s": 53.471846125030424, "throughput_rps": 0.2244171628550289, "latency_avg_s": 11.224783972332565, "latency_p50_s": 10.994812833494507, "latency_p95_s": 23.537262060036298, "latency_max_s": 25.672283249965403, "completion_tps": 16.158035725562083, "sample_errors": [], "peak_cpu_busy_pct": 55.35, "peak_load1": 7.28, "min_memory_free_pct": 14.0, "thermal_warning_seen": false, "performance_warning_seen": false, "thermal_status_peak": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded" }, "requests": [ { "request_id": "mac_mini-3-3", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138073.817645, "latency_s": 2.631130416993983, "finish_reason": "length", "content_preview": "- **Batch requests & use async I/O** \u2013 Process multiple incoming messages in parallel rather than one\u2011by\u2011one. \n- **Cach", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-3-1", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138073.81736, "latency_s": 4.251829000015277, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-3-4", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138076.448983, "latency_s": 2.189707166980952, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-3-2", "model": "qwen3:30b", "ok": true, "started_at": 1778138073.81756, "latency_s": 12.534756667038891, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-3-5", "model": "qwen3:30b", "ok": true, "started_at": 1778138078.06947, "latency_s": 9.89541337499395, "finish_reason": "length", "content_preview": "Okay, the user is testing concurrent chat capacity for a small AI inference cluster. They need three short operational t", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-3-6", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138078.6387699, "latency_s": 21.7904265410034, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages and process them in small, fixed-size batches to amortize model load and r", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-3-7", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138086.352475, "latency_s": 15.981095666997135, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-3-9", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138100.4293902, "latency_s": 3.273709000030067, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming queries and process them in small batches to amortize overhead. \n- **Scale horizon", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-3-10", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138102.333764, "latency_s": 10.467277624993585, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-3-8", "model": "qwen3:30b", "ok": true, "started_at": 1778138087.9650261, "latency_s": 25.672283249965403, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-3-11", "model": "qwen3:30b", "ok": true, "started_at": 1778138103.7032828, "latency_s": 11.522348041995429, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-3-12", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138112.801205, "latency_s": 14.487430916982703, "finish_reason": "length", "content_preview": "- **Batch requests & use async I/O** \u2013 Process multiple incoming messages in parallel and avoid blocking calls. \n- **Sc", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 } ], "samples": [ { "timestamp": 1778138073.611927, "unit": "mac_mini", "load1": 5.69, "load5": 5.79, "load15": 4.98, "cpu_user_pct": 31.12, "cpu_sys_pct": 10.5, "cpu_idle_pct": 58.36, "cpu_busy_pct": 41.620000000000005, "phys_mem_used": "47G", "phys_mem_unused": "638M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138076.883889, "unit": "mac_mini", "load1": 5.48, "load5": 5.75, "load15": 4.97, "cpu_user_pct": 23.37, "cpu_sys_pct": 19.91, "cpu_idle_pct": 56.7, "cpu_busy_pct": 43.28, "phys_mem_used": "47G", "phys_mem_unused": "383M", "memory_free_pct": 14.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138080.137766, "unit": "mac_mini", "load1": 5.48, "load5": 5.75, "load15": 4.97, "cpu_user_pct": 22.31, "cpu_sys_pct": 16.11, "cpu_idle_pct": 61.57, "cpu_busy_pct": 38.42, "phys_mem_used": "44G", "phys_mem_unused": "3406M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138083.396713, "unit": "mac_mini", "load1": 7.28, "load5": 6.12, "load15": 5.1, "cpu_user_pct": 21.61, "cpu_sys_pct": 20.33, "cpu_idle_pct": 58.5, "cpu_busy_pct": 41.94, "phys_mem_used": "46G", "phys_mem_unused": "1235M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138086.656096, "unit": "mac_mini", "load1": 7.02, "load5": 6.08, "load15": 5.1, "cpu_user_pct": 22.41, "cpu_sys_pct": 11.2, "cpu_idle_pct": 66.37, "cpu_busy_pct": 33.61, "phys_mem_used": "46G", "phys_mem_unused": "1522M", "memory_free_pct": 36.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138089.903391, "unit": "mac_mini", "load1": 7.02, "load5": 6.08, "load15": 5.1, "cpu_user_pct": 18.6, "cpu_sys_pct": 20.58, "cpu_idle_pct": 61.34, "cpu_busy_pct": 39.18, "phys_mem_used": "47G", "phys_mem_unused": "617M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138093.160774, "unit": "mac_mini", "load1": 7.02, "load5": 6.1, "load15": 5.11, "cpu_user_pct": 20.8, "cpu_sys_pct": 19.65, "cpu_idle_pct": 60.25, "cpu_busy_pct": 40.45, "phys_mem_used": "47G", "phys_mem_unused": "649M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138096.418628, "unit": "mac_mini", "load1": 6.94, "load5": 6.09, "load15": 5.11, "cpu_user_pct": 19.82, "cpu_sys_pct": 18.1, "cpu_idle_pct": 62.6, "cpu_busy_pct": 37.92, "phys_mem_used": "47G", "phys_mem_unused": "585M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138099.675811, "unit": "mac_mini", "load1": 6.94, "load5": 6.09, "load15": 5.11, "cpu_user_pct": 20.6, "cpu_sys_pct": 15.45, "cpu_idle_pct": 63.94, "cpu_busy_pct": 36.05, "phys_mem_used": "47G", "phys_mem_unused": "293M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138102.9294121, "unit": "mac_mini", "load1": 6.62, "load5": 6.04, "load15": 5.1, "cpu_user_pct": 30.4, "cpu_sys_pct": 12.87, "cpu_idle_pct": 57.8, "cpu_busy_pct": 43.269999999999996, "phys_mem_used": "46G", "phys_mem_unused": "1650M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138106.174461, "unit": "mac_mini", "load1": 6.65, "load5": 6.06, "load15": 5.11, "cpu_user_pct": 23.5, "cpu_sys_pct": 21.79, "cpu_idle_pct": 54.7, "cpu_busy_pct": 45.29, "phys_mem_used": "43G", "phys_mem_unused": "4883M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138109.43077, "unit": "mac_mini", "load1": 6.65, "load5": 6.06, "load15": 5.11, "cpu_user_pct": 20.83, "cpu_sys_pct": 20.41, "cpu_idle_pct": 58.75, "cpu_busy_pct": 41.239999999999995, "phys_mem_used": "45G", "phys_mem_unused": "2527M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138112.689749, "unit": "mac_mini", "load1": 6.36, "load5": 6.01, "load15": 5.1, "cpu_user_pct": 40.59, "cpu_sys_pct": 14.76, "cpu_idle_pct": 44.64, "cpu_busy_pct": 55.35, "phys_mem_used": "47G", "phys_mem_unused": "381M", "memory_free_pct": 25.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138115.980188, "unit": "mac_mini", "load1": 6.17, "load5": 5.97, "load15": 5.09, "cpu_user_pct": 16.66, "cpu_sys_pct": 18.37, "cpu_idle_pct": 64.95, "cpu_busy_pct": 35.03, "phys_mem_used": "47G", "phys_mem_unused": "542M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138119.243009, "unit": "mac_mini", "load1": 6.17, "load5": 5.97, "load15": 5.09, "cpu_user_pct": 13.98, "cpu_sys_pct": 13.55, "cpu_idle_pct": 72.45, "cpu_busy_pct": 27.53, "phys_mem_used": "47G", "phys_mem_unused": "679M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138122.50252, "unit": "mac_mini", "load1": 5.99, "load5": 5.94, "load15": 5.08, "cpu_user_pct": 20.94, "cpu_sys_pct": 17.9, "cpu_idle_pct": 61.96, "cpu_busy_pct": 38.84, "phys_mem_used": "47G", "phys_mem_unused": "575M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138125.757059, "unit": "mac_mini", "load1": 5.91, "load5": 5.93, "load15": 5.08, "cpu_user_pct": 30.21, "cpu_sys_pct": 14.46, "cpu_idle_pct": 55.31, "cpu_busy_pct": 44.67, "phys_mem_used": "47G", "phys_mem_unused": "525M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138127.490537, "unit": "mac_mini", "load1": 5.91, "load5": 5.93, "load15": 5.08, "cpu_user_pct": 14.6, "cpu_sys_pct": 10.17, "cpu_idle_pct": 75.22, "cpu_busy_pct": 24.77, "phys_mem_used": "47G", "phys_mem_unused": "485M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 } ] }, { "summary": { "concurrency": 4, "total_requests": 16, "request_count": 16, "success_count": 16, "error_count": 0, "error_rate": 0.0, "wall_s": 60.43576420901809, "throughput_rps": 0.26474390138699555, "latency_avg_s": 13.394817190190224, "latency_p50_s": 14.2997927495162, "latency_p95_s": 26.077366625278955, "latency_max_s": 28.597050375014078, "completion_tps": 18.267329195702693, "sample_errors": [], "peak_cpu_busy_pct": 44.67, "peak_load1": 5.91, "min_memory_free_pct": 15.0, "thermal_warning_seen": false, "performance_warning_seen": false, "thermal_status_peak": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded" }, "requests": [ { "request_id": "mac_mini-4-3", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138129.938297, "latency_s": 2.598843334009871, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages and process them in small batches to amortize model load and reduce per\u2011re", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-4-2", "model": "qwen3:30b", "ok": true, "started_at": 1778138129.938195, "latency_s": 11.972800584044307, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-4-1", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138129.937988, "latency_s": 13.661103916994762, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-4-4", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138129.9384372, "latency_s": 14.383132916002069, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-4-7", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138143.5993018, "latency_s": 1.426015458011534, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-4-5", "model": "qwen3:30b", "ok": true, "started_at": 1778138132.5373938, "latency_s": 13.13872408302268, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They want three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-4-8", "model": "qwen3:30b", "ok": true, "started_at": 1778138144.321704, "latency_s": 2.9520862499484792, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They need three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-4-6", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138141.911144, "latency_s": 17.468932166986633, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages into small batches before sending to the model; reduces per\u2011request overhe", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-4-10", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138145.676238, "latency_s": 15.055667916953098, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-4-9", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138145.025386, "latency_s": 17.544973791984376, "finish_reason": "length", "content_preview": "We need to give three short operational tips, concise bullet points. Provide short operational tips for keeping chat inf", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-4-12", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138159.380322, "latency_s": 5.734985000046436, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages and process them in small, fixed-size batches to amortize overhead. \n- **", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-4-13", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138160.7320619, "latency_s": 14.216452583030332, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 }, { "request_id": "mac_mini-4-11", "model": "qwen3:30b", "ok": true, "started_at": 1778138147.273941, "latency_s": 28.597050375014078, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster and wants three short operational tip", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-4-14", "model": "qwen3:30b", "ok": true, "started_at": 1778138162.5705862, "latency_s": 14.904496999981347, "finish_reason": "length", "content_preview": "Hmm, the user is testing concurrent chat capacity for a small AI inference cluster. They need three short operational ti", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "mac_mini-4-15", "model": "nemotron-3-nano:30b", "ok": true, "started_at": 1778138165.115391, "latency_s": 25.237472042033914, "finish_reason": "length", "content_preview": "- **Batch requests**: Group incoming messages and process them in small batches to amortize model loading overhead. \n- ", "prompt_tokens": 83, "completion_tokens": 96, "total_tokens": 179 }, { "request_id": "mac_mini-4-16", "model": "qwen2.5:7b", "ok": true, "started_at": 1778138174.948769, "latency_s": 15.424337624979671, "finish_reason": "stop", "content_preview": "- Optimize model efficiency with quantization.\n- Implement load balancing across nodes.\n- Use caching for frequent queri", "prompt_tokens": 77, "completion_tokens": 24, "total_tokens": 101 } ], "samples": [ { "timestamp": 1778138129.732574, "unit": "mac_mini", "load1": 5.91, "load5": 5.93, "load15": 5.08, "cpu_user_pct": 30.41, "cpu_sys_pct": 11.2, "cpu_idle_pct": 58.55, "cpu_busy_pct": 41.61, "phys_mem_used": "47G", "phys_mem_unused": "391M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138133.000232, "unit": "mac_mini", "load1": 5.68, "load5": 5.88, "load15": 5.07, "cpu_user_pct": 26.38, "cpu_sys_pct": 18.29, "cpu_idle_pct": 55.31, "cpu_busy_pct": 44.67, "phys_mem_used": "42G", "phys_mem_unused": "5734M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138136.245624, "unit": "mac_mini", "load1": 5.55, "load5": 5.85, "load15": 5.06, "cpu_user_pct": 24.12, "cpu_sys_pct": 18.85, "cpu_idle_pct": 57.1, "cpu_busy_pct": 42.97, "phys_mem_used": "45G", "phys_mem_unused": "2907M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138139.499784, "unit": "mac_mini", "load1": 5.55, "load5": 5.85, "load15": 5.06, "cpu_user_pct": 20.6, "cpu_sys_pct": 15.87, "cpu_idle_pct": 63.51, "cpu_busy_pct": 36.47, "phys_mem_used": "47G", "phys_mem_unused": "620M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138142.749534, "unit": "mac_mini", "load1": 5.5, "load5": 5.83, "load15": 5.06, "cpu_user_pct": 16.94, "cpu_sys_pct": 12.28, "cpu_idle_pct": 70.76, "cpu_busy_pct": 29.22, "phys_mem_used": "47G", "phys_mem_unused": "169M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138146.008429, "unit": "mac_mini", "load1": 5.54, "load5": 5.83, "load15": 5.07, "cpu_user_pct": 23.72, "cpu_sys_pct": 11.86, "cpu_idle_pct": 64.4, "cpu_busy_pct": 35.58, "phys_mem_used": "46G", "phys_mem_unused": "1611M", "memory_free_pct": 37.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138149.252915, "unit": "mac_mini", "load1": 5.54, "load5": 5.83, "load15": 5.07, "cpu_user_pct": 19.24, "cpu_sys_pct": 17.57, "cpu_idle_pct": 63.17, "cpu_busy_pct": 36.81, "phys_mem_used": "47G", "phys_mem_unused": "674M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138152.5054731, "unit": "mac_mini", "load1": 5.5, "load5": 5.82, "load15": 5.07, "cpu_user_pct": 19.32, "cpu_sys_pct": 18.48, "cpu_idle_pct": 62.18, "cpu_busy_pct": 37.8, "phys_mem_used": "47G", "phys_mem_unused": "585M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138155.7661939, "unit": "mac_mini", "load1": 5.54, "load5": 5.82, "load15": 5.07, "cpu_user_pct": 18.2, "cpu_sys_pct": 16.3, "cpu_idle_pct": 65.66, "cpu_busy_pct": 34.5, "phys_mem_used": "47G", "phys_mem_unused": "562M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138159.020288, "unit": "mac_mini", "load1": 5.54, "load5": 5.82, "load15": 5.07, "cpu_user_pct": 22.36, "cpu_sys_pct": 18.98, "cpu_idle_pct": 58.64, "cpu_busy_pct": 41.34, "phys_mem_used": "47G", "phys_mem_unused": "225M", "memory_free_pct": 17.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138162.278084, "unit": "mac_mini", "load1": 5.17, "load5": 5.74, "load15": 5.05, "cpu_user_pct": 13.27, "cpu_sys_pct": 16.59, "cpu_idle_pct": 70.12, "cpu_busy_pct": 29.86, "phys_mem_used": "46G", "phys_mem_unused": "1580M", "memory_free_pct": 26.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138165.534474, "unit": "mac_mini", "load1": 5.17, "load5": 5.74, "load15": 5.05, "cpu_user_pct": 13.86, "cpu_sys_pct": 20.58, "cpu_idle_pct": 65.54, "cpu_busy_pct": 34.44, "phys_mem_used": "41G", "phys_mem_unused": "6341M", "memory_free_pct": 38.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138168.77696, "unit": "mac_mini", "load1": 5.16, "load5": 5.73, "load15": 5.05, "cpu_user_pct": 10.37, "cpu_sys_pct": 17.84, "cpu_idle_pct": 71.78, "cpu_busy_pct": 28.21, "phys_mem_used": "43G", "phys_mem_unused": "4280M", "memory_free_pct": 38.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138172.047947, "unit": "mac_mini", "load1": 4.83, "load5": 5.65, "load15": 5.03, "cpu_user_pct": 12.39, "cpu_sys_pct": 17.76, "cpu_idle_pct": 69.83, "cpu_busy_pct": 30.150000000000002, "phys_mem_used": "45G", "phys_mem_unused": "2193M", "memory_free_pct": 38.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138175.318881, "unit": "mac_mini", "load1": 4.83, "load5": 5.65, "load15": 5.03, "cpu_user_pct": 14.63, "cpu_sys_pct": 15.44, "cpu_idle_pct": 69.91, "cpu_busy_pct": 30.07, "phys_mem_used": "46G", "phys_mem_unused": "1712M", "memory_free_pct": 38.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138178.582783, "unit": "mac_mini", "load1": 5.48, "load5": 5.77, "load15": 5.07, "cpu_user_pct": 9.34, "cpu_sys_pct": 17.47, "cpu_idle_pct": 73.17, "cpu_busy_pct": 26.81, "phys_mem_used": "47G", "phys_mem_unused": "572M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138181.853915, "unit": "mac_mini", "load1": 5.52, "load5": 5.78, "load15": 5.08, "cpu_user_pct": 12.89, "cpu_sys_pct": 19.92, "cpu_idle_pct": 67.18, "cpu_busy_pct": 32.81, "phys_mem_used": "47G", "phys_mem_unused": "653M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138185.1285539, "unit": "mac_mini", "load1": 5.52, "load5": 5.78, "load15": 5.08, "cpu_user_pct": 8.19, "cpu_sys_pct": 19.67, "cpu_idle_pct": 72.13, "cpu_busy_pct": 27.86, "phys_mem_used": "47G", "phys_mem_unused": "655M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138188.3986409, "unit": "mac_mini", "load1": 5.32, "load5": 5.73, "load15": 5.07, "cpu_user_pct": 11.15, "cpu_sys_pct": 13.63, "cpu_idle_pct": 75.2, "cpu_busy_pct": 24.78, "phys_mem_used": "47G", "phys_mem_unused": "485M", "memory_free_pct": 27.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 }, { "timestamp": 1778138190.578867, "unit": "mac_mini", "load1": 5.32, "load5": 5.73, "load15": 5.07, "cpu_user_pct": 13.33, "cpu_sys_pct": 14.66, "cpu_idle_pct": 72.0, "cpu_busy_pct": 27.990000000000002, "phys_mem_used": "47G", "phys_mem_unused": "139M", "memory_free_pct": 15.0, "thermal_status": "Note: No thermal warning level has been recorded\nNote: No performance warning level has been recorded\nNote: No CPU power status has been recorded", "thermal_warning": 0, "performance_warning": 0 } ] } ], "baseline_p50_s": 12.374412957986351 }, "spark": { "warm_results": [ { "request_id": "warm-spark-1", "model": "llava:latest", "ok": true, "started_at": 1778138874.220337, "latency_s": 3.691803375026211, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be inferred quickly to reduce latency.\n2. Implement efficient batch", "prompt_tokens": 75, "completion_tokens": 73, "total_tokens": 148 }, { "request_id": "warm-spark-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778138877.912492, "latency_s": 3.0711935000144877, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to reduce latency and improve throughput.\n* Optimi", "prompt_tokens": 76, "completion_tokens": 86, "total_tokens": 162 }, { "request_id": "warm-spark-3", "model": "nemotron:latest", "ok": true, "started_at": 1778138880.98399, "latency_s": 20.914444917056244, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "warm-spark-4", "model": "qwen3:8b", "ok": true, "started_at": 1778138901.898768, "latency_s": 2.504172917047981, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "steps": [ { "summary": { "concurrency": 1, "total_requests": 12, "request_count": 12, "success_count": 12, "error_count": 0, "error_rate": 0.0, "wall_s": 87.31058462499641, "throughput_rps": 0.13744038081454427, "latency_avg_s": 7.275378760251139, "latency_p50_s": 2.787085541029228, "latency_p95_s": 21.0708027437533, "latency_max_s": 21.083474125014618, "completion_tps": 10.663082878195059, "sample_errors": [], "peak_gpu_temp_c": 63.0, "peak_gpu_util_pct": 96.0, "peak_gpu_power_w": 49.61, "peak_load1": 1.23, "peak_mem_used_mb": 107560 }, "requests": [ { "request_id": "spark-1-1", "model": "llava:latest", "ok": true, "started_at": 1778138904.615563, "latency_s": 2.818743791023735, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency and increase throughput.\n2. Use efficient data structures and algo", "prompt_tokens": 75, "completion_tokens": 50, "total_tokens": 125 }, { "request_id": "spark-1-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778138907.434452, "latency_s": 2.8729636250063777, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests effici", "prompt_tokens": 76, "completion_tokens": 79, "total_tokens": 155 }, { "request_id": "spark-1-3", "model": "nemotron:latest", "ok": true, "started_at": 1778138910.307821, "latency_s": 20.997349124983884, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-1-4", "model": "qwen3:8b", "ok": true, "started_at": 1778138931.305357, "latency_s": 2.5408079589833505, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-1-5", "model": "llava:latest", "ok": true, "started_at": 1778138933.846577, "latency_s": 2.6527292910031974, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce latency\n2. Use efficient data structures for storing and processing chat messages\n3. I", "prompt_tokens": 75, "completion_tokens": 44, "total_tokens": 119 }, { "request_id": "spark-1-6", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778138936.499698, "latency_s": 2.8814695000182837, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models to reduce inference time, such as using model p", "prompt_tokens": 76, "completion_tokens": 76, "total_tokens": 152 }, { "request_id": "spark-1-7", "model": "nemotron:latest", "ok": true, "started_at": 1778138939.381391, "latency_s": 21.060435249994043, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-1-8", "model": "qwen3:8b", "ok": true, "started_at": 1778138960.4421172, "latency_s": 2.488563166989479, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-1-9", "model": "llava:latest", "ok": true, "started_at": 1778138962.9308228, "latency_s": 2.648017457977403, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce latency.\n2. Use efficient data structures and algorithms.\n3. Implement load balancing ", "prompt_tokens": 75, "completion_tokens": 41, "total_tokens": 116 }, { "request_id": "spark-1-10", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778138965.5792792, "latency_s": 2.755427291034721, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to minimize latency and improve throughput.\n* Para", "prompt_tokens": 76, "completion_tokens": 65, "total_tokens": 141 }, { "request_id": "spark-1-11", "model": "nemotron:latest", "ok": true, "started_at": 1778138968.335098, "latency_s": 21.083474125014618, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-1-12", "model": "qwen3:8b", "ok": true, "started_at": 1778138989.419244, "latency_s": 2.5045645409845747, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "samples": [ { "timestamp": 1778138904.4032729, "unit": "spark", "load1": 0.7, "load5": 0.9, "load15": 0.66, "mem_total_mb": 124610, "mem_used_mb": 97963, "mem_free_mb": 12398, "mem_available_mb": 26647, "gpu_temp_c": 52.0, "gpu_util_pct": 0.0, "gpu_power_w": 32.03 }, { "timestamp": 1778138907.936119, "unit": "spark", "load1": 0.96, "load5": 0.95, "load15": 0.68, "mem_total_mb": 124610, "mem_used_mb": 98394, "mem_free_mb": 11960, "mem_available_mb": 26216, "gpu_temp_c": 51.0, "gpu_util_pct": 9.0, "gpu_power_w": 17.49 }, { "timestamp": 1778138911.366452, "unit": "spark", "load1": 0.96, "load5": 0.95, "load15": 0.68, "mem_total_mb": 124610, "mem_used_mb": 101805, "mem_free_mb": 8549, "mem_available_mb": 22805, "gpu_temp_c": 55.0, "gpu_util_pct": 96.0, "gpu_power_w": 43.79 }, { "timestamp": 1778138914.822352, "unit": "spark", "load1": 0.97, "load5": 0.96, "load15": 0.68, "mem_total_mb": 124610, "mem_used_mb": 101798, "mem_free_mb": 8556, "mem_available_mb": 22812, "gpu_temp_c": 56.0, "gpu_util_pct": 96.0, "gpu_power_w": 43.98 }, { "timestamp": 1778138918.25027, "unit": "spark", "load1": 0.97, "load5": 0.96, "load15": 0.68, "mem_total_mb": 124610, "mem_used_mb": 101799, "mem_free_mb": 8554, "mem_available_mb": 22810, "gpu_temp_c": 57.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.08 }, { "timestamp": 1778138921.7207701, "unit": "spark", "load1": 0.97, "load5": 0.96, "load15": 0.68, "mem_total_mb": 124610, "mem_used_mb": 101805, "mem_free_mb": 8540, "mem_available_mb": 22805, "gpu_temp_c": 57.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.18 }, { "timestamp": 1778138925.520714, "unit": "spark", "load1": 1.05, "load5": 0.97, "load15": 0.69, "mem_total_mb": 124610, "mem_used_mb": 101808, "mem_free_mb": 8537, "mem_available_mb": 22802, "gpu_temp_c": 58.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.41 }, { "timestamp": 1778138928.954087, "unit": "spark", "load1": 1.05, "load5": 0.97, "load15": 0.69, "mem_total_mb": 124610, "mem_used_mb": 101816, "mem_free_mb": 8528, "mem_available_mb": 22794, "gpu_temp_c": 58.0, "gpu_util_pct": 84.0, "gpu_power_w": 43.21 }, { "timestamp": 1778138932.432585, "unit": "spark", "load1": 1.05, "load5": 0.97, "load15": 0.69, "mem_total_mb": 124610, "mem_used_mb": 101813, "mem_free_mb": 8530, "mem_available_mb": 22797, "gpu_temp_c": 59.0, "gpu_util_pct": 95.0, "gpu_power_w": 48.67 }, { "timestamp": 1778138935.9148521, "unit": "spark", "load1": 1.2, "load5": 1.01, "load15": 0.7, "mem_total_mb": 124610, "mem_used_mb": 107560, "mem_free_mb": 2782, "mem_available_mb": 17049, "gpu_temp_c": 58.0, "gpu_util_pct": 55.0, "gpu_power_w": 36.63 }, { "timestamp": 1778138939.3680398, "unit": "spark", "load1": 1.11, "load5": 0.99, "load15": 0.7, "mem_total_mb": 124610, "mem_used_mb": 101830, "mem_free_mb": 8512, "mem_available_mb": 22779, "gpu_temp_c": 58.0, "gpu_util_pct": 89.0, "gpu_power_w": 42.22 }, { "timestamp": 1778138942.768329, "unit": "spark", "load1": 1.11, "load5": 0.99, "load15": 0.7, "mem_total_mb": 124610, "mem_used_mb": 101821, "mem_free_mb": 8522, "mem_available_mb": 22789, "gpu_temp_c": 59.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.62 }, { "timestamp": 1778138946.205833, "unit": "spark", "load1": 1.18, "load5": 1.01, "load15": 0.71, "mem_total_mb": 124610, "mem_used_mb": 101813, "mem_free_mb": 8530, "mem_available_mb": 22797, "gpu_temp_c": 59.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.57 }, { "timestamp": 1778138949.623188, "unit": "spark", "load1": 1.16, "load5": 1.01, "load15": 0.71, "mem_total_mb": 124610, "mem_used_mb": 101828, "mem_free_mb": 8515, "mem_available_mb": 22782, "gpu_temp_c": 60.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.74 }, { "timestamp": 1778138953.121953, "unit": "spark", "load1": 1.23, "load5": 1.02, "load15": 0.71, "mem_total_mb": 124610, "mem_used_mb": 101824, "mem_free_mb": 8518, "mem_available_mb": 22786, "gpu_temp_c": 60.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.71 }, { "timestamp": 1778138956.63794, "unit": "spark", "load1": 1.23, "load5": 1.02, "load15": 0.71, "mem_total_mb": 124610, "mem_used_mb": 101816, "mem_free_mb": 8526, "mem_available_mb": 22794, "gpu_temp_c": 61.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.2 }, { "timestamp": 1778138960.082371, "unit": "spark", "load1": 1.21, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101826, "mem_free_mb": 8514, "mem_available_mb": 22783, "gpu_temp_c": 61.0, "gpu_util_pct": 96.0, "gpu_power_w": 43.81 }, { "timestamp": 1778138963.486294, "unit": "spark", "load1": 1.2, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 102323, "mem_free_mb": 8018, "mem_available_mb": 22287, "gpu_temp_c": 57.0, "gpu_util_pct": 9.0, "gpu_power_w": 19.6 }, { "timestamp": 1778138966.87781, "unit": "spark", "load1": 1.2, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101809, "mem_free_mb": 8531, "mem_available_mb": 22801, "gpu_temp_c": 56.0, "gpu_util_pct": 6.0, "gpu_power_w": 15.29 }, { "timestamp": 1778138970.226849, "unit": "spark", "load1": 1.18, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101827, "mem_free_mb": 8513, "mem_available_mb": 22782, "gpu_temp_c": 60.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.04 }, { "timestamp": 1778138973.681639, "unit": "spark", "load1": 1.16, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101820, "mem_free_mb": 8520, "mem_available_mb": 22790, "gpu_temp_c": 61.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.0 }, { "timestamp": 1778138977.0943801, "unit": "spark", "load1": 1.16, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101822, "mem_free_mb": 8519, "mem_available_mb": 22788, "gpu_temp_c": 61.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.34 }, { "timestamp": 1778138980.5814888, "unit": "spark", "load1": 1.15, "load5": 1.02, "load15": 0.72, "mem_total_mb": 124610, "mem_used_mb": 101813, "mem_free_mb": 8526, "mem_available_mb": 22796, "gpu_temp_c": 61.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.52 }, { "timestamp": 1778138983.9977758, "unit": "spark", "load1": 1.14, "load5": 1.02, "load15": 0.73, "mem_total_mb": 124610, "mem_used_mb": 101805, "mem_free_mb": 8535, "mem_available_mb": 22805, "gpu_temp_c": 62.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.44 }, { "timestamp": 1778138987.709131, "unit": "spark", "load1": 1.14, "load5": 1.02, "load15": 0.73, "mem_total_mb": 124610, "mem_used_mb": 101801, "mem_free_mb": 8539, "mem_available_mb": 22809, "gpu_temp_c": 62.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.61 }, { "timestamp": 1778138991.180181, "unit": "spark", "load1": 1.21, "load5": 1.04, "load15": 0.73, "mem_total_mb": 124610, "mem_used_mb": 101790, "mem_free_mb": 8550, "mem_available_mb": 22820, "gpu_temp_c": 63.0, "gpu_util_pct": 95.0, "gpu_power_w": 49.61 }, { "timestamp": 1778138992.13164, "unit": "spark", "load1": 1.21, "load5": 1.04, "load15": 0.73, "mem_total_mb": 124610, "mem_used_mb": 101788, "mem_free_mb": 8552, "mem_available_mb": 22822, "gpu_temp_c": 58.0, "gpu_util_pct": 95.0, "gpu_power_w": 32.23 } ] }, { "summary": { "concurrency": 2, "total_requests": 12, "request_count": 12, "success_count": 12, "error_count": 0, "error_rate": 0.0, "wall_s": 88.69139995798469, "throughput_rps": 0.13530060418129264, "latency_avg_s": 13.330229774330897, "latency_p50_s": 9.189531667012488, "latency_p95_s": 30.284897798232848, "latency_max_s": 31.567722541978583, "completion_tps": 11.737327412727137, "sample_errors": [], "peak_gpu_temp_c": 66.0, "peak_gpu_util_pct": 96.0, "peak_gpu_power_w": 46.61, "peak_load1": 3.62, "peak_mem_used_mb": 107717 }, "requests": [ { "request_id": "spark-2-1", "model": "llava:latest", "ok": true, "started_at": 1778138994.6842098, "latency_s": 3.424049791006837, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models or prune larger ones to reduce memory usage and latency.\n\n2. Implement effic", "prompt_tokens": 75, "completion_tokens": 70, "total_tokens": 145 }, { "request_id": "spark-2-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778138994.6843672, "latency_s": 9.565843084012158, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can process data quickly, reducing respons", "prompt_tokens": 76, "completion_tokens": 70, "total_tokens": 146 }, { "request_id": "spark-2-4", "model": "qwen3:8b", "ok": true, "started_at": 1778139004.250339, "latency_s": 12.213246792030986, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-2-5", "model": "llava:latest", "ok": true, "started_at": 1778139016.46399, "latency_s": 7.811381875013467, "finish_reason": "stop", "content_preview": " 1. Optimize the AI model for efficient inference to reduce latency.\n2. Implement load balancing to distribute incoming ", "prompt_tokens": 75, "completion_tokens": 52, "total_tokens": 127 }, { "request_id": "spark-2-3", "model": "nemotron:latest", "ok": true, "started_at": 1778138998.1087308, "latency_s": 29.235313916986343, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-2-6", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139024.275819, "latency_s": 7.897505333006848, "finish_reason": "stop", "content_preview": " - Use efficient algorithms: Implement optimized inference models that can handle multiple requests simultaneously, such", "prompt_tokens": 76, "completion_tokens": 93, "total_tokens": 169 }, { "request_id": "spark-2-8", "model": "qwen3:8b", "ok": true, "started_at": 1778139032.173782, "latency_s": 7.062651874963194, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-2-9", "model": "llava:latest", "ok": true, "started_at": 1778139039.236971, "latency_s": 9.995615582971368, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be inferred quickly, reducing the latency of responses.\n2. Implemen", "prompt_tokens": 75, "completion_tokens": 84, "total_tokens": 159 }, { "request_id": "spark-2-10", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139049.2330391, "latency_s": 8.813220250012819, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests quickl", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-2-7", "model": "nemotron:latest", "ok": true, "started_at": 1778139027.344557, "latency_s": 31.567722541978583, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-2-12", "model": "qwen3:8b", "ok": true, "started_at": 1778139058.912607, "latency_s": 7.0477377499919385, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-2-11", "model": "nemotron:latest", "ok": true, "started_at": 1778139058.046526, "latency_s": 25.328468499996234, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "samples": [ { "timestamp": 1778138994.471599, "unit": "spark", "load1": 1.27, "load5": 1.05, "load15": 0.74, "mem_total_mb": 124610, "mem_used_mb": 101798, "mem_free_mb": 8541, "mem_available_mb": 22812, "gpu_temp_c": 57.0, "gpu_util_pct": 0.0, "gpu_power_w": 14.85 }, { "timestamp": 1778138997.81117, "unit": "spark", "load1": 1.27, "load5": 1.05, "load15": 0.74, "mem_total_mb": 124610, "mem_used_mb": 96084, "mem_free_mb": 14257, "mem_available_mb": 28525, "gpu_temp_c": 60.0, "gpu_util_pct": 95.0, "gpu_power_w": 38.47 }, { "timestamp": 1778139001.2433019, "unit": "spark", "load1": 1.65, "load5": 1.14, "load15": 0.77, "mem_total_mb": 124610, "mem_used_mb": 99635, "mem_free_mb": 10704, "mem_available_mb": 24975, "gpu_temp_c": 62.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.12 }, { "timestamp": 1778139004.6962, "unit": "spark", "load1": 1.68, "load5": 1.15, "load15": 0.78, "mem_total_mb": 124610, "mem_used_mb": 99238, "mem_free_mb": 11101, "mem_available_mb": 25372, "gpu_temp_c": 62.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.09 }, { "timestamp": 1778139008.319351, "unit": "spark", "load1": 1.7, "load5": 1.17, "load15": 0.78, "mem_total_mb": 124610, "mem_used_mb": 101711, "mem_free_mb": 8920, "mem_available_mb": 22899, "gpu_temp_c": 62.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.0 }, { "timestamp": 1778139011.857145, "unit": "spark", "load1": 1.7, "load5": 1.17, "load15": 0.78, "mem_total_mb": 124610, "mem_used_mb": 101960, "mem_free_mb": 8671, "mem_available_mb": 22650, "gpu_temp_c": 63.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.16 }, { "timestamp": 1778139015.2675672, "unit": "spark", "load1": 1.73, "load5": 1.18, "load15": 0.79, "mem_total_mb": 124610, "mem_used_mb": 101988, "mem_free_mb": 8643, "mem_available_mb": 22622, "gpu_temp_c": 63.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.32 }, { "timestamp": 1778139018.767255, "unit": "spark", "load1": 1.83, "load5": 1.21, "load15": 0.8, "mem_total_mb": 124610, "mem_used_mb": 102461, "mem_free_mb": 8170, "mem_available_mb": 22149, "gpu_temp_c": 63.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.07 }, { "timestamp": 1778139022.18968, "unit": "spark", "load1": 1.83, "load5": 1.21, "load15": 0.8, "mem_total_mb": 124610, "mem_used_mb": 107717, "mem_free_mb": 2913, "mem_available_mb": 16892, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.07 }, { "timestamp": 1778139025.565114, "unit": "spark", "load1": 1.92, "load5": 1.24, "load15": 0.81, "mem_total_mb": 124610, "mem_used_mb": 98357, "mem_free_mb": 12273, "mem_available_mb": 26253, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.38 }, { "timestamp": 1778139029.001819, "unit": "spark", "load1": 1.93, "load5": 1.25, "load15": 0.82, "mem_total_mb": 124610, "mem_used_mb": 101989, "mem_free_mb": 8641, "mem_available_mb": 22621, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.03 }, { "timestamp": 1778139032.405018, "unit": "spark", "load1": 1.93, "load5": 1.25, "load15": 0.82, "mem_total_mb": 124610, "mem_used_mb": 101962, "mem_free_mb": 8667, "mem_available_mb": 22647, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.58 }, { "timestamp": 1778139035.8169868, "unit": "spark", "load1": 3.62, "load5": 1.61, "load15": 0.94, "mem_total_mb": 124610, "mem_used_mb": 101940, "mem_free_mb": 8667, "mem_available_mb": 22669, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.72 }, { "timestamp": 1778139039.242545, "unit": "spark", "load1": 3.57, "load5": 1.64, "load15": 0.95, "mem_total_mb": 124610, "mem_used_mb": 101664, "mem_free_mb": 8943, "mem_available_mb": 22946, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.52 }, { "timestamp": 1778139042.944156, "unit": "spark", "load1": 3.6, "load5": 1.68, "load15": 0.97, "mem_total_mb": 124610, "mem_used_mb": 107547, "mem_free_mb": 3353, "mem_available_mb": 17063, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.92 }, { "timestamp": 1778139046.395947, "unit": "spark", "load1": 3.6, "load5": 1.68, "load15": 0.97, "mem_total_mb": 124610, "mem_used_mb": 107660, "mem_free_mb": 3233, "mem_available_mb": 16950, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.54 }, { "timestamp": 1778139050.220781, "unit": "spark", "load1": 3.47, "load5": 1.68, "load15": 0.97, "mem_total_mb": 124610, "mem_used_mb": 98153, "mem_free_mb": 12740, "mem_available_mb": 26457, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.42 }, { "timestamp": 1778139053.622127, "unit": "spark", "load1": 3.36, "load5": 1.69, "load15": 0.98, "mem_total_mb": 124610, "mem_used_mb": 101939, "mem_free_mb": 8953, "mem_available_mb": 22671, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.05 }, { "timestamp": 1778139057.034428, "unit": "spark", "load1": 3.36, "load5": 1.69, "load15": 0.98, "mem_total_mb": 124610, "mem_used_mb": 101940, "mem_free_mb": 8952, "mem_available_mb": 22670, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.49 }, { "timestamp": 1778139060.525919, "unit": "spark", "load1": 3.25, "load5": 1.69, "load15": 0.99, "mem_total_mb": 124610, "mem_used_mb": 101936, "mem_free_mb": 8956, "mem_available_mb": 22673, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.1 }, { "timestamp": 1778139063.95607, "unit": "spark", "load1": 3.15, "load5": 1.7, "load15": 0.99, "mem_total_mb": 124610, "mem_used_mb": 101945, "mem_free_mb": 8948, "mem_available_mb": 22665, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.04 }, { "timestamp": 1778139067.3620028, "unit": "spark", "load1": 3.15, "load5": 1.7, "load15": 0.99, "mem_total_mb": 124610, "mem_used_mb": 101937, "mem_free_mb": 8956, "mem_available_mb": 22673, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139070.7901728, "unit": "spark", "load1": 3.06, "load5": 1.7, "load15": 1.0, "mem_total_mb": 124610, "mem_used_mb": 101956, "mem_free_mb": 8936, "mem_available_mb": 22654, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.46 }, { "timestamp": 1778139074.280235, "unit": "spark", "load1": 2.89, "load5": 1.69, "load15": 1.0, "mem_total_mb": 124610, "mem_used_mb": 101940, "mem_free_mb": 8952, "mem_available_mb": 22670, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.49 }, { "timestamp": 1778139077.760362, "unit": "spark", "load1": 2.89, "load5": 1.69, "load15": 1.0, "mem_total_mb": 124610, "mem_used_mb": 101949, "mem_free_mb": 8943, "mem_available_mb": 22661, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.42 }, { "timestamp": 1778139081.23623, "unit": "spark", "load1": 2.98, "load5": 1.73, "load15": 1.01, "mem_total_mb": 124610, "mem_used_mb": 101945, "mem_free_mb": 8946, "mem_available_mb": 22665, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.41 }, { "timestamp": 1778139083.581696, "unit": "spark", "load1": 2.82, "load5": 1.72, "load15": 1.01, "mem_total_mb": 124610, "mem_used_mb": 101935, "mem_free_mb": 8956, "mem_available_mb": 22675, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 30.09 } ] }, { "summary": { "concurrency": 4, "total_requests": 16, "request_count": 16, "success_count": 16, "error_count": 0, "error_rate": 0.0, "wall_s": 115.54268729197793, "throughput_rps": 0.138476959252019, "latency_avg_s": 22.310767362047045, "latency_p50_s": 11.786476020992268, "latency_p95_s": 59.26891944797535, "latency_max_s": 62.25069016701309, "completion_tps": 11.926328115580136, "sample_errors": [], "peak_gpu_temp_c": 65.0, "peak_gpu_util_pct": 96.0, "peak_gpu_power_w": 48.01, "peak_load1": 3.17, "peak_mem_used_mb": 107654 }, "requests": [ { "request_id": "spark-4-4", "model": "qwen3:8b", "ok": true, "started_at": 1778139086.158352, "latency_s": 8.299695915949997, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-1", "model": "llava:latest", "ok": true, "started_at": 1778139086.157821, "latency_s": 11.126016874972265, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be loaded faster, reducing the latency of responses.\n2. Implement l", "prompt_tokens": 75, "completion_tokens": 84, "total_tokens": 159 }, { "request_id": "spark-4-5", "model": "llava:latest", "ok": true, "started_at": 1778139094.458316, "latency_s": 7.248635499970987, "finish_reason": "stop", "content_preview": " 1. Optimize the model: Use efficient models that are optimized for inference to reduce latency.\n2. Load balance: Distri", "prompt_tokens": 75, "completion_tokens": 74, "total_tokens": 149 }, { "request_id": "spark-4-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139086.158036, "latency_s": 22.299941167002544, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to minimize latency and improve throughput.\n* Para", "prompt_tokens": 76, "completion_tokens": 62, "total_tokens": 138 }, { "request_id": "spark-4-6", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139097.28419, "latency_s": 16.676338125020266, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference models that minimize latency and improve throughput, such as ", "prompt_tokens": 76, "completion_tokens": 87, "total_tokens": 163 }, { "request_id": "spark-4-8", "model": "qwen3:8b", "ok": true, "started_at": 1778139108.4585228, "latency_s": 8.527736791991629, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-9", "model": "llava:latest", "ok": true, "started_at": 1778139113.96084, "latency_s": 8.497868999955244, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency and increase throughput.\n2. Use efficient data structures and algo", "prompt_tokens": 75, "completion_tokens": 55, "total_tokens": 130 }, { "request_id": "spark-4-3", "model": "nemotron:latest", "ok": true, "started_at": 1778139086.158185, "latency_s": 38.995147667010315, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-10", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139116.9865868, "latency_s": 11.26212025003042, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests effici", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-4-12", "model": "qwen3:8b", "ok": true, "started_at": 1778139125.153885, "latency_s": 13.02645212499192, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-13", "model": "llava:latest", "ok": true, "started_at": 1778139128.249259, "latency_s": 11.19887274998473, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce memory usage and improve inference speed.\n2. Use efficient data structures and algorit", "prompt_tokens": 75, "completion_tokens": 56, "total_tokens": 131 }, { "request_id": "spark-4-14", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139138.180649, "latency_s": 9.434529458987527, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests quickl", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-4-7", "model": "nemotron:latest", "ok": true, "started_at": 1778139101.707396, "latency_s": 57.5424043329549, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across multiple AI nodes\n\t+ U", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-16", "model": "qwen3:8b", "ok": true, "started_at": 1778139147.615691, "latency_s": 12.310831791954115, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-11", "model": "nemotron:latest", "ok": true, "started_at": 1778139122.459011, "latency_s": 58.27499587496277, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balance**: \n\t+ Distribute incoming chats across multiple AI nodes\n\t+ Us", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-4-15", "model": "nemotron:latest", "ok": true, "started_at": 1778139139.448576, "latency_s": 62.25069016701309, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "samples": [ { "timestamp": 1778139085.9472241, "unit": "spark", "load1": 2.82, "load5": 1.72, "load15": 1.01, "mem_total_mb": 124610, "mem_used_mb": 101930, "mem_free_mb": 8961, "mem_available_mb": 22679, "gpu_temp_c": 59.0, "gpu_util_pct": 0.0, "gpu_power_w": 15.22 }, { "timestamp": 1778139089.2797298, "unit": "spark", "load1": 2.84, "load5": 1.74, "load15": 1.02, "mem_total_mb": 124610, "mem_used_mb": 106868, "mem_free_mb": 4316, "mem_available_mb": 17742, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.1 }, { "timestamp": 1778139092.773406, "unit": "spark", "load1": 2.84, "load5": 1.74, "load15": 1.02, "mem_total_mb": 124610, "mem_used_mb": 107642, "mem_free_mb": 3541, "mem_available_mb": 16968, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.83 }, { "timestamp": 1778139096.217921, "unit": "spark", "load1": 3.17, "load5": 1.83, "load15": 1.06, "mem_total_mb": 124610, "mem_used_mb": 107648, "mem_free_mb": 3535, "mem_available_mb": 16962, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.19 }, { "timestamp": 1778139099.676312, "unit": "spark", "load1": 3.08, "load5": 1.83, "load15": 1.06, "mem_total_mb": 124610, "mem_used_mb": 107654, "mem_free_mb": 3529, "mem_available_mb": 16956, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.24 }, { "timestamp": 1778139103.10156, "unit": "spark", "load1": 2.99, "load5": 1.83, "load15": 1.07, "mem_total_mb": 124610, "mem_used_mb": 98369, "mem_free_mb": 12813, "mem_available_mb": 26241, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.61 }, { "timestamp": 1778139106.5893881, "unit": "spark", "load1": 2.99, "load5": 1.83, "load15": 1.07, "mem_total_mb": 124610, "mem_used_mb": 101989, "mem_free_mb": 9192, "mem_available_mb": 22620, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.04 }, { "timestamp": 1778139110.00369, "unit": "spark", "load1": 2.91, "load5": 1.84, "load15": 1.07, "mem_total_mb": 124610, "mem_used_mb": 102023, "mem_free_mb": 9158, "mem_available_mb": 22587, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.15 }, { "timestamp": 1778139113.694922, "unit": "spark", "load1": 2.92, "load5": 1.86, "load15": 1.08, "mem_total_mb": 124610, "mem_used_mb": 102011, "mem_free_mb": 9170, "mem_available_mb": 22599, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.9 }, { "timestamp": 1778139117.2161, "unit": "spark", "load1": 2.92, "load5": 1.86, "load15": 1.08, "mem_total_mb": 124610, "mem_used_mb": 106841, "mem_free_mb": 4632, "mem_available_mb": 17768, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.38 }, { "timestamp": 1778139120.686563, "unit": "spark", "load1": 2.92, "load5": 1.87, "load15": 1.09, "mem_total_mb": 124610, "mem_used_mb": 96099, "mem_free_mb": 15374, "mem_available_mb": 28511, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 41.82 }, { "timestamp": 1778139124.15419, "unit": "spark", "load1": 2.85, "load5": 1.88, "load15": 1.1, "mem_total_mb": 124610, "mem_used_mb": 99693, "mem_free_mb": 12073, "mem_available_mb": 24917, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.33 }, { "timestamp": 1778139127.6036139, "unit": "spark", "load1": 2.85, "load5": 1.88, "load15": 1.1, "mem_total_mb": 124610, "mem_used_mb": 90845, "mem_free_mb": 20920, "mem_available_mb": 33765, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.31 }, { "timestamp": 1778139131.330719, "unit": "spark", "load1": 3.02, "load5": 1.93, "load15": 1.12, "mem_total_mb": 124610, "mem_used_mb": 102543, "mem_free_mb": 9510, "mem_available_mb": 22067, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.43 }, { "timestamp": 1778139134.73699, "unit": "spark", "load1": 3.02, "load5": 1.95, "load15": 1.13, "mem_total_mb": 124610, "mem_used_mb": 107595, "mem_free_mb": 4679, "mem_available_mb": 17015, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.27 }, { "timestamp": 1778139138.195488, "unit": "spark", "load1": 3.1, "load5": 1.98, "load15": 1.15, "mem_total_mb": 124610, "mem_used_mb": 107647, "mem_free_mb": 4626, "mem_available_mb": 16963, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.7 }, { "timestamp": 1778139141.603918, "unit": "spark", "load1": 3.1, "load5": 1.98, "load15": 1.15, "mem_total_mb": 124610, "mem_used_mb": 98583, "mem_free_mb": 13816, "mem_available_mb": 26027, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 48.01 }, { "timestamp": 1778139145.07073, "unit": "spark", "load1": 2.93, "load5": 1.96, "load15": 1.14, "mem_total_mb": 124610, "mem_used_mb": 99780, "mem_free_mb": 12617, "mem_available_mb": 24830, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.35 }, { "timestamp": 1778139148.484291, "unit": "spark", "load1": 2.94, "load5": 1.98, "load15": 1.16, "mem_total_mb": 124610, "mem_used_mb": 90407, "mem_free_mb": 21966, "mem_available_mb": 34203, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.18 }, { "timestamp": 1778139152.030207, "unit": "spark", "load1": 2.94, "load5": 1.98, "load15": 1.16, "mem_total_mb": 124610, "mem_used_mb": 102039, "mem_free_mb": 10344, "mem_available_mb": 22571, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.42 }, { "timestamp": 1778139155.637675, "unit": "spark", "load1": 2.94, "load5": 2.0, "load15": 1.17, "mem_total_mb": 124610, "mem_used_mb": 102078, "mem_free_mb": 10302, "mem_available_mb": 22532, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.85 }, { "timestamp": 1778139159.058965, "unit": "spark", "load1": 2.87, "load5": 2.0, "load15": 1.17, "mem_total_mb": 124610, "mem_used_mb": 102090, "mem_free_mb": 10290, "mem_available_mb": 22520, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.92 }, { "timestamp": 1778139162.499986, "unit": "spark", "load1": 2.87, "load5": 2.0, "load15": 1.17, "mem_total_mb": 124610, "mem_used_mb": 102069, "mem_free_mb": 10310, "mem_available_mb": 22540, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.83 }, { "timestamp": 1778139166.0092149, "unit": "spark", "load1": 3.12, "load5": 2.07, "load15": 1.2, "mem_total_mb": 124610, "mem_used_mb": 102076, "mem_free_mb": 10302, "mem_available_mb": 22533, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.67 }, { "timestamp": 1778139169.4651039, "unit": "spark", "load1": 2.95, "load5": 2.05, "load15": 1.2, "mem_total_mb": 124610, "mem_used_mb": 102075, "mem_free_mb": 10302, "mem_available_mb": 22535, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.43 }, { "timestamp": 1778139172.98998, "unit": "spark", "load1": 2.95, "load5": 2.06, "load15": 1.2, "mem_total_mb": 124610, "mem_used_mb": 102058, "mem_free_mb": 10308, "mem_available_mb": 22552, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.83 }, { "timestamp": 1778139176.8436742, "unit": "spark", "load1": 2.95, "load5": 2.06, "load15": 1.2, "mem_total_mb": 124610, "mem_used_mb": 102072, "mem_free_mb": 10292, "mem_available_mb": 22537, "gpu_temp_c": 64.0, "gpu_util_pct": 84.0, "gpu_power_w": 45.56 }, { "timestamp": 1778139180.2763572, "unit": "spark", "load1": 3.04, "load5": 2.1, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 102059, "mem_free_mb": 10304, "mem_available_mb": 22551, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.21 }, { "timestamp": 1778139183.736062, "unit": "spark", "load1": 2.87, "load5": 2.08, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 102021, "mem_free_mb": 10342, "mem_available_mb": 22589, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.31 }, { "timestamp": 1778139187.1709812, "unit": "spark", "load1": 2.87, "load5": 2.08, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 102004, "mem_free_mb": 10359, "mem_available_mb": 22606, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.28 }, { "timestamp": 1778139190.615123, "unit": "spark", "load1": 2.72, "load5": 2.06, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 102156, "mem_free_mb": 9983, "mem_available_mb": 22454, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.08 }, { "timestamp": 1778139194.081024, "unit": "spark", "load1": 2.66, "load5": 2.06, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 102221, "mem_free_mb": 9739, "mem_available_mb": 22389, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.29 }, { "timestamp": 1778139197.518642, "unit": "spark", "load1": 2.66, "load5": 2.06, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 101987, "mem_free_mb": 9973, "mem_available_mb": 22623, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.3 }, { "timestamp": 1778139200.9764879, "unit": "spark", "load1": 2.61, "load5": 2.06, "load15": 1.23, "mem_total_mb": 124610, "mem_used_mb": 101981, "mem_free_mb": 9979, "mem_available_mb": 22629, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.33 }, { "timestamp": 1778139201.903109, "unit": "spark", "load1": 2.61, "load5": 2.06, "load15": 1.23, "mem_total_mb": 124610, "mem_used_mb": 101973, "mem_free_mb": 9987, "mem_available_mb": 22637, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 31.3 } ] }, { "summary": { "concurrency": 6, "total_requests": 24, "request_count": 24, "success_count": 24, "error_count": 0, "error_rate": 0.0, "wall_s": 168.23583712504478, "throughput_rps": 0.14265688220852435, "latency_avg_s": 29.826767461750325, "latency_p50_s": 15.215512667025905, "latency_p95_s": 95.34877841250852, "latency_max_s": 100.9161638750229, "completion_tps": 11.691920304340307, "sample_errors": [], "peak_gpu_temp_c": 66.0, "peak_gpu_util_pct": 96.0, "peak_gpu_power_w": 47.52, "peak_load1": 3.07, "peak_mem_used_mb": 107876 }, "requests": [ { "request_id": "spark-6-4", "model": "qwen3:8b", "ok": true, "started_at": 1778139204.4369438, "latency_s": 8.22675283299759, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-5", "model": "llava:latest", "ok": true, "started_at": 1778139204.437065, "latency_s": 9.320884833985474, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency and increase throughput.\n2. Use efficient data structures and algo", "prompt_tokens": 75, "completion_tokens": 54, "total_tokens": 129 }, { "request_id": "spark-6-1", "model": "llava:latest", "ok": true, "started_at": 1778139204.4364328, "latency_s": 16.457281958020758, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be inferred quickly to reduce latency.\n\n2. Implement load balancing", "prompt_tokens": 75, "completion_tokens": 83, "total_tokens": 158 }, { "request_id": "spark-6-8", "model": "qwen3:8b", "ok": true, "started_at": 1778139213.758052, "latency_s": 9.067571708001196, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139204.436651, "latency_s": 23.682719000033103, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to minimize latency and improve throughput.\n* Para", "prompt_tokens": 76, "completion_tokens": 70, "total_tokens": 146 }, { "request_id": "spark-6-6", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139204.437187, "latency_s": 28.62696170801064, "finish_reason": "length", "content_preview": " - Use efficient algorithms: Implement parallel processing techniques to optimize inference speed, such as using GPU acc", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-6-9", "model": "llava:latest", "ok": true, "started_at": 1778139220.894102, "latency_s": 16.36492745799478, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be inferred quickly, reducing latency.\n2. Implement load balancing:", "prompt_tokens": 75, "completion_tokens": 77, "total_tokens": 152 }, { "request_id": "spark-6-10", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139222.825929, "latency_s": 15.331141041999217, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference models that minimize latency and improve throughput, such as ", "prompt_tokens": 76, "completion_tokens": 85, "total_tokens": 161 }, { "request_id": "spark-6-14", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139238.157349, "latency_s": 3.9907612079987302, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference models that minimize latency, such as using techniques like m", "prompt_tokens": 76, "completion_tokens": 83, "total_tokens": 159 }, { "request_id": "spark-6-12", "model": "qwen3:8b", "ok": true, "started_at": 1778139233.064638, "latency_s": 17.33518950000871, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-3", "model": "nemotron:latest", "ok": true, "started_at": 1778139204.4368, "latency_s": 46.22128195798723, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **1. Caching**:\n\t+ Cache frequent user queries/results\n\t+ Reduce repeated infe", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-13", "model": "llava:latest", "ok": true, "started_at": 1778139237.259317, "latency_s": 15.099884292052593, "finish_reason": "stop", "content_preview": " 1. Optimize the AI model to reduce latency and increase throughput.\n2. Implement load balancing to distribute user requ", "prompt_tokens": 75, "completion_tokens": 52, "total_tokens": 127 }, { "request_id": "spark-6-17", "model": "llava:latest", "ok": true, "started_at": 1778139250.658397, "latency_s": 4.502562249952462, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce latency\n2. Use efficient data structures and algorithms\n3. Implement load balancing an", "prompt_tokens": 75, "completion_tokens": 32, "total_tokens": 107 }, { "request_id": "spark-6-16", "model": "qwen3:8b", "ok": true, "started_at": 1778139250.400121, "latency_s": 8.516498207987752, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-18", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139252.359622, "latency_s": 11.494084291975014, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to minimize latency and improve throughput.\n* Opti", "prompt_tokens": 76, "completion_tokens": 73, "total_tokens": 149 }, { "request_id": "spark-6-20", "model": "qwen3:8b", "ok": true, "started_at": 1778139258.91682, "latency_s": 8.477670499996748, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-21", "model": "llava:latest", "ok": true, "started_at": 1778139263.8552, "latency_s": 7.899294457980432, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency.\n2. Use efficient data structures and algorithms for processing in", "prompt_tokens": 75, "completion_tokens": 47, "total_tokens": 122 }, { "request_id": "spark-6-22", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139267.3949282, "latency_s": 9.170121957955416, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to minimize latency and improve throughput.\n* Para", "prompt_tokens": 76, "completion_tokens": 63, "total_tokens": 139 }, { "request_id": "spark-6-24", "model": "qwen3:8b", "ok": true, "started_at": 1778139276.5654988, "latency_s": 12.081169125041924, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-7", "model": "nemotron:latest", "ok": true, "started_at": 1778139212.6641269, "latency_s": 76.23139345803065, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-11", "model": "nemotron:latest", "ok": true, "started_at": 1778139228.1199, "latency_s": 81.69047495897394, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-15", "model": "nemotron:latest", "ok": true, "started_at": 1778139242.148472, "latency_s": 88.59743687498849, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-19", "model": "nemotron:latest", "ok": true, "started_at": 1778139255.161367, "latency_s": 96.54019162501208, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-6-23", "model": "nemotron:latest", "ok": true, "started_at": 1778139271.755114, "latency_s": 100.9161638750229, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "samples": [ { "timestamp": 1778139204.226275, "unit": "spark", "load1": 2.4, "load5": 2.02, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 101946, "mem_free_mb": 10013, "mem_available_mb": 22663, "gpu_temp_c": 59.0, "gpu_util_pct": 0.0, "gpu_power_w": 15.21 }, { "timestamp": 1778139207.587648, "unit": "spark", "load1": 2.4, "load5": 2.02, "load15": 1.22, "mem_total_mb": 124610, "mem_used_mb": 107077, "mem_free_mb": 5145, "mem_available_mb": 17533, "gpu_temp_c": 63.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.32 }, { "timestamp": 1778139211.1430988, "unit": "spark", "load1": 2.53, "load5": 2.06, "load15": 1.23, "mem_total_mb": 124610, "mem_used_mb": 107811, "mem_free_mb": 4406, "mem_available_mb": 16798, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.83 }, { "timestamp": 1778139214.644274, "unit": "spark", "load1": 2.49, "load5": 2.06, "load15": 1.24, "mem_total_mb": 124610, "mem_used_mb": 107796, "mem_free_mb": 4421, "mem_available_mb": 16813, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.97 }, { "timestamp": 1778139218.06948, "unit": "spark", "load1": 2.53, "load5": 2.07, "load15": 1.25, "mem_total_mb": 124610, "mem_used_mb": 107795, "mem_free_mb": 4422, "mem_available_mb": 16815, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.97 }, { "timestamp": 1778139221.626382, "unit": "spark", "load1": 2.53, "load5": 2.07, "load15": 1.25, "mem_total_mb": 124610, "mem_used_mb": 98314, "mem_free_mb": 13903, "mem_available_mb": 26296, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.06 }, { "timestamp": 1778139225.1201699, "unit": "spark", "load1": 2.49, "load5": 2.07, "load15": 1.25, "mem_total_mb": 124610, "mem_used_mb": 101318, "mem_free_mb": 10898, "mem_available_mb": 23292, "gpu_temp_c": 64.0, "gpu_util_pct": 95.0, "gpu_power_w": 43.96 }, { "timestamp": 1778139228.848437, "unit": "spark", "load1": 2.77, "load5": 2.14, "load15": 1.28, "mem_total_mb": 124610, "mem_used_mb": 98710, "mem_free_mb": 14092, "mem_available_mb": 25900, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.92 }, { "timestamp": 1778139232.371349, "unit": "spark", "load1": 2.77, "load5": 2.14, "load15": 1.28, "mem_total_mb": 124610, "mem_used_mb": 99800, "mem_free_mb": 13001, "mem_available_mb": 24809, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.32 }, { "timestamp": 1778139235.872245, "unit": "spark", "load1": 2.87, "load5": 2.17, "load15": 1.29, "mem_total_mb": 124610, "mem_used_mb": 99832, "mem_free_mb": 12961, "mem_available_mb": 24777, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.32 }, { "timestamp": 1778139239.64039, "unit": "spark", "load1": 2.88, "load5": 2.18, "load15": 1.3, "mem_total_mb": 124610, "mem_used_mb": 92496, "mem_free_mb": 20584, "mem_available_mb": 32114, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.6 }, { "timestamp": 1778139243.521017, "unit": "spark", "load1": 2.89, "load5": 2.19, "load15": 1.31, "mem_total_mb": 124610, "mem_used_mb": 99145, "mem_free_mb": 13098, "mem_available_mb": 25465, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.12 }, { "timestamp": 1778139247.320329, "unit": "spark", "load1": 2.89, "load5": 2.19, "load15": 1.31, "mem_total_mb": 124610, "mem_used_mb": 107855, "mem_free_mb": 4451, "mem_available_mb": 16755, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.47 }, { "timestamp": 1778139250.825122, "unit": "spark", "load1": 2.9, "load5": 2.21, "load15": 1.32, "mem_total_mb": 124610, "mem_used_mb": 107848, "mem_free_mb": 4458, "mem_available_mb": 16762, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.24 }, { "timestamp": 1778139254.279116, "unit": "spark", "load1": 2.91, "load5": 2.22, "load15": 1.33, "mem_total_mb": 124610, "mem_used_mb": 107856, "mem_free_mb": 4449, "mem_available_mb": 16754, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.2 }, { "timestamp": 1778139257.717925, "unit": "spark", "load1": 2.91, "load5": 2.22, "load15": 1.33, "mem_total_mb": 124610, "mem_used_mb": 100960, "mem_free_mb": 11345, "mem_available_mb": 23650, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.28 }, { "timestamp": 1778139261.167216, "unit": "spark", "load1": 3.07, "load5": 2.27, "load15": 1.35, "mem_total_mb": 124610, "mem_used_mb": 102153, "mem_free_mb": 10152, "mem_available_mb": 22457, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139264.586098, "unit": "spark", "load1": 3.07, "load5": 2.28, "load15": 1.36, "mem_total_mb": 124610, "mem_used_mb": 98577, "mem_free_mb": 13727, "mem_available_mb": 26032, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.4 }, { "timestamp": 1778139268.114055, "unit": "spark", "load1": 2.98, "load5": 2.28, "load15": 1.36, "mem_total_mb": 124610, "mem_used_mb": 107876, "mem_free_mb": 4461, "mem_available_mb": 16733, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.19 }, { "timestamp": 1778139271.683063, "unit": "spark", "load1": 2.98, "load5": 2.28, "load15": 1.36, "mem_total_mb": 124610, "mem_used_mb": 98627, "mem_free_mb": 13740, "mem_available_mb": 25982, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.69 }, { "timestamp": 1778139275.372098, "unit": "spark", "load1": 3.06, "load5": 2.3, "load15": 1.38, "mem_total_mb": 124610, "mem_used_mb": 99839, "mem_free_mb": 12525, "mem_available_mb": 24770, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.41 }, { "timestamp": 1778139278.8391838, "unit": "spark", "load1": 3.06, "load5": 2.32, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 90882, "mem_free_mb": 21482, "mem_available_mb": 33728, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 47.52 }, { "timestamp": 1778139282.309406, "unit": "spark", "load1": 3.06, "load5": 2.32, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102275, "mem_free_mb": 10335, "mem_available_mb": 22335, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 47.02 }, { "timestamp": 1778139285.766819, "unit": "spark", "load1": 3.05, "load5": 2.33, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102195, "mem_free_mb": 10415, "mem_available_mb": 22415, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.19 }, { "timestamp": 1778139289.17109, "unit": "spark", "load1": 2.97, "load5": 2.32, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102189, "mem_free_mb": 10419, "mem_available_mb": 22421, "gpu_temp_c": 66.0, "gpu_util_pct": 94.0, "gpu_power_w": 46.49 }, { "timestamp": 1778139292.64627, "unit": "spark", "load1": 2.97, "load5": 2.32, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102187, "mem_free_mb": 10420, "mem_available_mb": 22422, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.71 }, { "timestamp": 1778139296.0592482, "unit": "spark", "load1": 2.89, "load5": 2.32, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102186, "mem_free_mb": 10421, "mem_available_mb": 22424, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.67 }, { "timestamp": 1778139299.534226, "unit": "spark", "load1": 2.74, "load5": 2.29, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102178, "mem_free_mb": 10419, "mem_available_mb": 22432, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.44 }, { "timestamp": 1778139303.4365911, "unit": "spark", "load1": 2.6, "load5": 2.27, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102155, "mem_free_mb": 10441, "mem_available_mb": 22454, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.57 }, { "timestamp": 1778139306.861297, "unit": "spark", "load1": 2.6, "load5": 2.27, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102144, "mem_free_mb": 10452, "mem_available_mb": 22466, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.51 }, { "timestamp": 1778139310.2653599, "unit": "spark", "load1": 2.47, "load5": 2.25, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102134, "mem_free_mb": 10462, "mem_available_mb": 22476, "gpu_temp_c": 65.0, "gpu_util_pct": 85.0, "gpu_power_w": 46.28 }, { "timestamp": 1778139313.6785111, "unit": "spark", "load1": 2.35, "load5": 2.23, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102121, "mem_free_mb": 10475, "mem_available_mb": 22489, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139317.098805, "unit": "spark", "load1": 2.35, "load5": 2.23, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102140, "mem_free_mb": 10379, "mem_available_mb": 22470, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.45 }, { "timestamp": 1778139320.521396, "unit": "spark", "load1": 2.24, "load5": 2.21, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102112, "mem_free_mb": 10406, "mem_available_mb": 22497, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.52 }, { "timestamp": 1778139323.970164, "unit": "spark", "load1": 2.14, "load5": 2.19, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102106, "mem_free_mb": 10413, "mem_available_mb": 22504, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.59 }, { "timestamp": 1778139327.410656, "unit": "spark", "load1": 2.14, "load5": 2.19, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102092, "mem_free_mb": 10426, "mem_available_mb": 22518, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.46 }, { "timestamp": 1778139330.876788, "unit": "spark", "load1": 2.21, "load5": 2.2, "load15": 1.4, "mem_total_mb": 124610, "mem_used_mb": 102097, "mem_free_mb": 10421, "mem_available_mb": 22512, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.37 }, { "timestamp": 1778139334.3215559, "unit": "spark", "load1": 2.12, "load5": 2.18, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102092, "mem_free_mb": 10425, "mem_available_mb": 22517, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.48 }, { "timestamp": 1778139337.802613, "unit": "spark", "load1": 2.12, "load5": 2.18, "load15": 1.39, "mem_total_mb": 124610, "mem_used_mb": 102089, "mem_free_mb": 10429, "mem_available_mb": 22521, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.54 }, { "timestamp": 1778139341.254456, "unit": "spark", "load1": 2.27, "load5": 2.21, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102098, "mem_free_mb": 10420, "mem_available_mb": 22512, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.49 }, { "timestamp": 1778139344.725881, "unit": "spark", "load1": 2.33, "load5": 2.23, "load15": 1.42, "mem_total_mb": 124610, "mem_used_mb": 102144, "mem_free_mb": 10373, "mem_available_mb": 22465, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.54 }, { "timestamp": 1778139348.22107, "unit": "spark", "load1": 2.22, "load5": 2.21, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102129, "mem_free_mb": 10388, "mem_available_mb": 22481, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139351.68828, "unit": "spark", "load1": 2.22, "load5": 2.21, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102118, "mem_free_mb": 10399, "mem_available_mb": 22492, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.41 }, { "timestamp": 1778139355.151076, "unit": "spark", "load1": 2.12, "load5": 2.19, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102106, "mem_free_mb": 10411, "mem_available_mb": 22504, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139358.6212149, "unit": "spark", "load1": 2.03, "load5": 2.17, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102106, "mem_free_mb": 10412, "mem_available_mb": 22504, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.53 }, { "timestamp": 1778139362.05972, "unit": "spark", "load1": 2.03, "load5": 2.17, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102112, "mem_free_mb": 10405, "mem_available_mb": 22498, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.44 }, { "timestamp": 1778139365.823484, "unit": "spark", "load1": 2.03, "load5": 2.16, "load15": 1.41, "mem_total_mb": 124610, "mem_used_mb": 102275, "mem_free_mb": 10002, "mem_available_mb": 22334, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.42 }, { "timestamp": 1778139369.235022, "unit": "spark", "load1": 2.03, "load5": 2.16, "load15": 1.42, "mem_total_mb": 124610, "mem_used_mb": 102093, "mem_free_mb": 10016, "mem_available_mb": 22517, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.4 }, { "timestamp": 1778139372.643181, "unit": "spark", "load1": 2.03, "load5": 2.16, "load15": 1.42, "mem_total_mb": 124610, "mem_used_mb": 102110, "mem_free_mb": 9999, "mem_available_mb": 22499, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 38.0 }, { "timestamp": 1778139372.981717, "unit": "spark", "load1": 2.03, "load5": 2.16, "load15": 1.42, "mem_total_mb": 124610, "mem_used_mb": 102133, "mem_free_mb": 9976, "mem_available_mb": 22476, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 27.76 } ] }, { "summary": { "concurrency": 8, "total_requests": 32, "request_count": 32, "success_count": 32, "error_count": 0, "error_rate": 0.0, "wall_s": 223.116339124972, "throughput_rps": 0.14342293408675977, "latency_avg_s": 38.14707736721721, "latency_p50_s": 17.78267337498255, "latency_p95_s": 127.34639058297907, "latency_max_s": 134.57571875001304, "completion_tps": 12.1192379303312, "sample_errors": [], "peak_gpu_temp_c": 67.0, "peak_gpu_util_pct": 96.0, "peak_gpu_power_w": 47.16, "peak_load1": 2.89, "peak_mem_used_mb": 108065 }, "requests": [ { "request_id": "spark-8-4", "model": "qwen3:8b", "ok": true, "started_at": 1778139375.525961, "latency_s": 8.19357524998486, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need operational ", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-5", "model": "llava:latest", "ok": true, "started_at": 1778139375.526082, "latency_s": 8.456236708036158, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce latency\n2. Use efficient data structures for input processing\n3. Implement load balanc", "prompt_tokens": 75, "completion_tokens": 38, "total_tokens": 113 }, { "request_id": "spark-8-1", "model": "llava:latest", "ok": true, "started_at": 1778139375.525429, "latency_s": 15.933707499993034, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models that can be loaded and inferred quickly, reducing latency.\n2. Implement load", "prompt_tokens": 75, "completion_tokens": 85, "total_tokens": 160 }, { "request_id": "spark-8-8", "model": "qwen3:8b", "ok": true, "started_at": 1778139375.5264509, "latency_s": 17.662233916984405, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-9", "model": "llava:latest", "ok": true, "started_at": 1778139383.719904, "latency_s": 11.74643937498331, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency.\n2. Use efficient data structures and algorithms for processing in", "prompt_tokens": 75, "completion_tokens": 46, "total_tokens": 121 }, { "request_id": "spark-8-12", "model": "qwen3:8b", "ok": true, "started_at": 1778139393.1889439, "latency_s": 8.204366165969986, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-2", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139375.525641, "latency_s": 28.9993622500333, "finish_reason": "length", "content_preview": " - Use efficient algorithms: Implement optimized inference models that can handle high volumes of requests efficiently, ", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-8-6", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139375.526218, "latency_s": 33.181354959029704, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests quickl", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-8-10", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139383.982616, "latency_s": 30.301839750027284, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference models that can handle multiple requests concurrently, ensuri", "prompt_tokens": 76, "completion_tokens": 87, "total_tokens": 163 }, { "request_id": "spark-8-16", "model": "qwen3:8b", "ok": true, "started_at": 1778139408.7078888, "latency_s": 9.121572416974232, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-14", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139401.393738, "latency_s": 17.390362624952104, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models to reduce inference time, such as using model p", "prompt_tokens": 76, "completion_tokens": 75, "total_tokens": 151 }, { "request_id": "spark-8-18", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139417.829939, "latency_s": 5.12307262502145, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized machine learning models that can handle high volumes of requests effici", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-8-7", "model": "nemotron:latest", "ok": true, "started_at": 1778139375.5263438, "latency_s": 48.00194529205328, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-20", "model": "qwen3:8b", "ok": true, "started_at": 1778139422.953421, "latency_s": 8.542301458015572, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-13", "model": "llava:latest", "ok": true, "started_at": 1778139395.466708, "latency_s": 36.95754087500973, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency.\n2. Use efficient data structures and algorithms for processing in", "prompt_tokens": 75, "completion_tokens": 46, "total_tokens": 121 }, { "request_id": "spark-8-17", "model": "llava:latest", "ok": true, "started_at": 1778139414.2848792, "latency_s": 20.66066504199989, "finish_reason": "stop", "content_preview": " 1. Optimize model size to reduce latency\n2. Use efficient data structures for input and output\n3. Implement load balanc", "prompt_tokens": 75, "completion_tokens": 39, "total_tokens": 114 }, { "request_id": "spark-8-21", "model": "llava:latest", "ok": true, "started_at": 1778139423.528781, "latency_s": 17.903112832980696, "finish_reason": "stop", "content_preview": " 1. Optimize model size: Use smaller models or prune larger ones to reduce memory usage and latency.\n\n2. Implement load ", "prompt_tokens": 75, "completion_tokens": 85, "total_tokens": 160 }, { "request_id": "spark-8-22", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139431.49599, "latency_s": 10.81818491697777, "finish_reason": "length", "content_preview": " * Use efficient algorithms: Implement optimized inference models that can handle high volumes of requests quickly, such", "prompt_tokens": 76, "completion_tokens": 96, "total_tokens": 172 }, { "request_id": "spark-8-26", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139442.3144681, "latency_s": 4.078389416972641, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference models that minimize latency, such as using techniques like m", "prompt_tokens": 76, "completion_tokens": 85, "total_tokens": 161 }, { "request_id": "spark-8-24", "model": "qwen3:8b", "ok": true, "started_at": 1778139434.9460502, "latency_s": 19.581673624983523, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-25", "model": "llava:latest", "ok": true, "started_at": 1778139441.432354, "latency_s": 18.067364874994382, "finish_reason": "stop", "content_preview": " 1. Optimize the model: Use efficient models that are optimized for inference to reduce latency.\n\n2. Distribute workload", "prompt_tokens": 75, "completion_tokens": 88, "total_tokens": 163 }, { "request_id": "spark-8-29", "model": "llava:latest", "ok": true, "started_at": 1778139459.500064, "latency_s": 4.127396833966486, "finish_reason": "stop", "content_preview": " 1. Optimize the model architecture to reduce latency and memory usage.\n2. Use efficient data structures and algorithms ", "prompt_tokens": 75, "completion_tokens": 47, "total_tokens": 122 }, { "request_id": "spark-8-28", "model": "qwen3:8b", "ok": true, "started_at": 1778139454.528102, "latency_s": 9.535531750007067, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-3", "model": "nemotron:latest", "ok": true, "started_at": 1778139375.5257988, "latency_s": 92.78141204098938, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-30", "model": "nemotron-mini:latest", "ok": true, "started_at": 1778139463.627889, "latency_s": 8.107702250010334, "finish_reason": "stop", "content_preview": " * Use efficient algorithms: Implement optimized inference algorithms to reduce latency and improve throughput.\n* Optimi", "prompt_tokens": 76, "completion_tokens": 63, "total_tokens": 139 }, { "request_id": "spark-8-32", "model": "qwen3:8b", "ok": true, "started_at": 1778139468.3080058, "latency_s": 7.866115290962625, "finish_reason": "length", "content_preview": "Okay, the user is testing their AI inference cluster's ability to handle concurrent chat traffic. They need three operat", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-11", "model": "nemotron:latest", "ok": true, "started_at": 1778139391.459568, "latency_s": 102.30593383405358, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across multiple AI nodes\n\t+ E", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-15", "model": "nemotron:latest", "ok": true, "started_at": 1778139404.525561, "latency_s": 110.18720004201168, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-19", "model": "nemotron:latest", "ok": true, "started_at": 1778139418.784279, "latency_s": 116.90575291699497, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-23", "model": "nemotron:latest", "ok": true, "started_at": 1778139432.424665, "latency_s": 124.21606008295203, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-27", "model": "nemotron:latest", "ok": true, "started_at": 1778139446.393257, "latency_s": 131.17235008301213, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balancing**:\n\t+ Distribute incoming chats across all available AI nodes", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 }, { "request_id": "spark-8-31", "model": "nemotron:latest", "ok": true, "started_at": 1778139464.0638962, "latency_s": 134.57571875001304, "finish_reason": "length", "content_preview": "Here are three short operational tips:\n\n* **Load Balance**: \n\t+ Distribute incoming chats across multiple AI nodes\n\t+ Us", "prompt_tokens": 79, "completion_tokens": 96, "total_tokens": 175 } ], "samples": [ { "timestamp": 1778139375.315008, "unit": "spark", "load1": 2.02, "load5": 2.16, "load15": 1.42, "mem_total_mb": 124610, "mem_used_mb": 102203, "mem_free_mb": 9906, "mem_available_mb": 22406, "gpu_temp_c": 59.0, "gpu_util_pct": 0.0, "gpu_power_w": 15.32 }, { "timestamp": 1778139378.681098, "unit": "spark", "load1": 2.1, "load5": 2.17, "load15": 1.43, "mem_total_mb": 124610, "mem_used_mb": 107791, "mem_free_mb": 4580, "mem_available_mb": 16819, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.57 }, { "timestamp": 1778139382.371125, "unit": "spark", "load1": 2.1, "load5": 2.17, "load15": 1.43, "mem_total_mb": 124610, "mem_used_mb": 107931, "mem_free_mb": 4438, "mem_available_mb": 16678, "gpu_temp_c": 64.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.94 }, { "timestamp": 1778139385.811508, "unit": "spark", "load1": 2.17, "load5": 2.18, "load15": 1.44, "mem_total_mb": 124610, "mem_used_mb": 107894, "mem_free_mb": 4475, "mem_available_mb": 16715, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.08 }, { "timestamp": 1778139389.2471948, "unit": "spark", "load1": 2.24, "load5": 2.2, "load15": 1.45, "mem_total_mb": 124610, "mem_used_mb": 107897, "mem_free_mb": 4472, "mem_available_mb": 16712, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.16 }, { "timestamp": 1778139392.748733, "unit": "spark", "load1": 2.24, "load5": 2.2, "load15": 1.45, "mem_total_mb": 124610, "mem_used_mb": 107885, "mem_free_mb": 4484, "mem_available_mb": 16725, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.17 }, { "timestamp": 1778139396.231446, "unit": "spark", "load1": 2.38, "load5": 2.23, "load15": 1.46, "mem_total_mb": 124610, "mem_used_mb": 98487, "mem_free_mb": 13882, "mem_available_mb": 26123, "gpu_temp_c": 65.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.44 }, { "timestamp": 1778139399.7399218, "unit": "spark", "load1": 2.51, "load5": 2.26, "load15": 1.47, "mem_total_mb": 124610, "mem_used_mb": 102216, "mem_free_mb": 10153, "mem_available_mb": 22394, "gpu_temp_c": 66.0, "gpu_util_pct": 95.0, "gpu_power_w": 46.22 }, { "timestamp": 1778139403.2879071, "unit": "spark", "load1": 2.47, "load5": 2.25, "load15": 1.48, "mem_total_mb": 124610, "mem_used_mb": 102204, "mem_free_mb": 10164, "mem_available_mb": 22405, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.33 }, { "timestamp": 1778139406.829151, "unit": "spark", "load1": 2.47, "load5": 2.25, "load15": 1.48, "mem_total_mb": 124610, "mem_used_mb": 102216, "mem_free_mb": 10153, "mem_available_mb": 22394, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.34 }, { "timestamp": 1778139410.28667, "unit": "spark", "load1": 2.43, "load5": 2.25, "load15": 1.48, "mem_total_mb": 124610, "mem_used_mb": 102213, "mem_free_mb": 10156, "mem_available_mb": 22397, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.59 }, { "timestamp": 1778139413.6703382, "unit": "spark", "load1": 2.48, "load5": 2.26, "load15": 1.49, "mem_total_mb": 124610, "mem_used_mb": 102222, "mem_free_mb": 10147, "mem_available_mb": 22388, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.54 }, { "timestamp": 1778139417.101304, "unit": "spark", "load1": 2.48, "load5": 2.26, "load15": 1.49, "mem_total_mb": 124610, "mem_used_mb": 102215, "mem_free_mb": 10153, "mem_available_mb": 22394, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.72 }, { "timestamp": 1778139420.5223758, "unit": "spark", "load1": 2.44, "load5": 2.26, "load15": 1.49, "mem_total_mb": 124610, "mem_used_mb": 102215, "mem_free_mb": 10153, "mem_available_mb": 22395, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.47 }, { "timestamp": 1778139423.992705, "unit": "spark", "load1": 2.4, "load5": 2.25, "load15": 1.49, "mem_total_mb": 124610, "mem_used_mb": 102203, "mem_free_mb": 10161, "mem_available_mb": 22406, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.57 }, { "timestamp": 1778139427.91623, "unit": "spark", "load1": 2.61, "load5": 2.3, "load15": 1.51, "mem_total_mb": 124610, "mem_used_mb": 107965, "mem_free_mb": 4692, "mem_available_mb": 16645, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.69 }, { "timestamp": 1778139431.399693, "unit": "spark", "load1": 2.61, "load5": 2.3, "load15": 1.51, "mem_total_mb": 124610, "mem_used_mb": 107937, "mem_free_mb": 4720, "mem_available_mb": 16673, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.16 }, { "timestamp": 1778139434.816488, "unit": "spark", "load1": 2.64, "load5": 2.31, "load15": 1.52, "mem_total_mb": 124610, "mem_used_mb": 98805, "mem_free_mb": 14145, "mem_available_mb": 25805, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.71 }, { "timestamp": 1778139438.411059, "unit": "spark", "load1": 2.67, "load5": 2.32, "load15": 1.53, "mem_total_mb": 124610, "mem_used_mb": 99982, "mem_free_mb": 12967, "mem_available_mb": 24628, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.69 }, { "timestamp": 1778139441.826922, "unit": "spark", "load1": 2.67, "load5": 2.32, "load15": 1.53, "mem_total_mb": 124610, "mem_used_mb": 90476, "mem_free_mb": 22473, "mem_available_mb": 34134, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 44.77 }, { "timestamp": 1778139445.243146, "unit": "spark", "load1": 2.7, "load5": 2.33, "load15": 1.54, "mem_total_mb": 124610, "mem_used_mb": 102036, "mem_free_mb": 11195, "mem_available_mb": 22574, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.03 }, { "timestamp": 1778139448.756814, "unit": "spark", "load1": 2.72, "load5": 2.35, "load15": 1.55, "mem_total_mb": 124610, "mem_used_mb": 102884, "mem_free_mb": 9557, "mem_available_mb": 21726, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.25 }, { "timestamp": 1778139452.154963, "unit": "spark", "load1": 2.72, "load5": 2.35, "load15": 1.55, "mem_total_mb": 124610, "mem_used_mb": 108059, "mem_free_mb": 4675, "mem_available_mb": 16550, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.53 }, { "timestamp": 1778139455.6066718, "unit": "spark", "load1": 2.75, "load5": 2.36, "load15": 1.55, "mem_total_mb": 124610, "mem_used_mb": 108062, "mem_free_mb": 4671, "mem_available_mb": 16548, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.55 }, { "timestamp": 1778139459.086387, "unit": "spark", "load1": 2.77, "load5": 2.37, "load15": 1.56, "mem_total_mb": 124610, "mem_used_mb": 108059, "mem_free_mb": 4674, "mem_available_mb": 16551, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.6 }, { "timestamp": 1778139462.4986901, "unit": "spark", "load1": 2.77, "load5": 2.37, "load15": 1.56, "mem_total_mb": 124610, "mem_used_mb": 108065, "mem_free_mb": 4668, "mem_available_mb": 16545, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.62 }, { "timestamp": 1778139465.94861, "unit": "spark", "load1": 2.87, "load5": 2.4, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 99014, "mem_free_mb": 13719, "mem_available_mb": 25596, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 47.16 }, { "timestamp": 1778139469.5824041, "unit": "spark", "load1": 2.8, "load5": 2.39, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102338, "mem_free_mb": 10394, "mem_available_mb": 22271, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.95 }, { "timestamp": 1778139473.006334, "unit": "spark", "load1": 2.89, "load5": 2.42, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102337, "mem_free_mb": 10395, "mem_available_mb": 22272, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.39 }, { "timestamp": 1778139476.4388359, "unit": "spark", "load1": 2.89, "load5": 2.42, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102322, "mem_free_mb": 10410, "mem_available_mb": 22287, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.71 }, { "timestamp": 1778139479.841443, "unit": "spark", "load1": 2.74, "load5": 2.39, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102316, "mem_free_mb": 10416, "mem_available_mb": 22294, "gpu_temp_c": 67.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.13 }, { "timestamp": 1778139483.2821481, "unit": "spark", "load1": 2.6, "load5": 2.37, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102309, "mem_free_mb": 10423, "mem_available_mb": 22301, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.47 }, { "timestamp": 1778139486.7664099, "unit": "spark", "load1": 2.6, "load5": 2.37, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102307, "mem_free_mb": 10413, "mem_available_mb": 22302, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.31 }, { "timestamp": 1778139490.5785, "unit": "spark", "load1": 2.47, "load5": 2.35, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102308, "mem_free_mb": 10411, "mem_available_mb": 22302, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.67 }, { "timestamp": 1778139494.0072808, "unit": "spark", "load1": 2.35, "load5": 2.32, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102309, "mem_free_mb": 10410, "mem_available_mb": 22301, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.74 }, { "timestamp": 1778139497.4562418, "unit": "spark", "load1": 2.35, "load5": 2.32, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102312, "mem_free_mb": 10407, "mem_available_mb": 22298, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.72 }, { "timestamp": 1778139500.8792078, "unit": "spark", "load1": 2.33, "load5": 2.32, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102322, "mem_free_mb": 10397, "mem_available_mb": 22288, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.65 }, { "timestamp": 1778139504.3038359, "unit": "spark", "load1": 2.22, "load5": 2.3, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102317, "mem_free_mb": 10402, "mem_available_mb": 22293, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139507.854649, "unit": "spark", "load1": 2.22, "load5": 2.3, "load15": 1.58, "mem_total_mb": 124610, "mem_used_mb": 102308, "mem_free_mb": 10410, "mem_available_mb": 22302, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139511.307949, "unit": "spark", "load1": 2.6, "load5": 2.37, "load15": 1.61, "mem_total_mb": 124610, "mem_used_mb": 102344, "mem_free_mb": 10374, "mem_available_mb": 22266, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139514.698165, "unit": "spark", "load1": 2.47, "load5": 2.35, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102336, "mem_free_mb": 10383, "mem_available_mb": 22274, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.27 }, { "timestamp": 1778139518.158599, "unit": "spark", "load1": 2.36, "load5": 2.33, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102324, "mem_free_mb": 10394, "mem_available_mb": 22285, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.62 }, { "timestamp": 1778139521.6901262, "unit": "spark", "load1": 2.36, "load5": 2.33, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102371, "mem_free_mb": 10296, "mem_available_mb": 22238, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.02 }, { "timestamp": 1778139525.157374, "unit": "spark", "load1": 2.33, "load5": 2.32, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102309, "mem_free_mb": 10062, "mem_available_mb": 22300, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.44 }, { "timestamp": 1778139528.6358888, "unit": "spark", "load1": 2.22, "load5": 2.3, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102303, "mem_free_mb": 10068, "mem_available_mb": 22306, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.6 }, { "timestamp": 1778139532.104645, "unit": "spark", "load1": 2.22, "load5": 2.3, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102315, "mem_free_mb": 10057, "mem_available_mb": 22295, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.68 }, { "timestamp": 1778139535.565656, "unit": "spark", "load1": 2.12, "load5": 2.28, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102309, "mem_free_mb": 10062, "mem_available_mb": 22301, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.55 }, { "timestamp": 1778139539.1092281, "unit": "spark", "load1": 2.03, "load5": 2.26, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102306, "mem_free_mb": 10065, "mem_available_mb": 22304, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.56 }, { "timestamp": 1778139542.654599, "unit": "spark", "load1": 2.03, "load5": 2.26, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102322, "mem_free_mb": 10049, "mem_available_mb": 22288, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139546.133784, "unit": "spark", "load1": 2.11, "load5": 2.27, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102324, "mem_free_mb": 10047, "mem_available_mb": 22285, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.64 }, { "timestamp": 1778139549.565763, "unit": "spark", "load1": 2.02, "load5": 2.25, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102307, "mem_free_mb": 10063, "mem_available_mb": 22302, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.54 }, { "timestamp": 1778139553.3022048, "unit": "spark", "load1": 2.1, "load5": 2.26, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102309, "mem_free_mb": 10062, "mem_available_mb": 22301, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.58 }, { "timestamp": 1778139556.811543, "unit": "spark", "load1": 2.1, "load5": 2.26, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102315, "mem_free_mb": 10056, "mem_available_mb": 22295, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.93 }, { "timestamp": 1778139560.263294, "unit": "spark", "load1": 2.01, "load5": 2.24, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102310, "mem_free_mb": 10061, "mem_available_mb": 22300, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.57 }, { "timestamp": 1778139563.721102, "unit": "spark", "load1": 1.93, "load5": 2.22, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102308, "mem_free_mb": 10063, "mem_available_mb": 22302, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.64 }, { "timestamp": 1778139567.303204, "unit": "spark", "load1": 1.93, "load5": 2.22, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102305, "mem_free_mb": 10066, "mem_available_mb": 22305, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.62 }, { "timestamp": 1778139570.9036689, "unit": "spark", "load1": 1.94, "load5": 2.21, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102303, "mem_free_mb": 10068, "mem_available_mb": 22307, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.6 }, { "timestamp": 1778139574.387112, "unit": "spark", "load1": 1.86, "load5": 2.19, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102311, "mem_free_mb": 10060, "mem_available_mb": 22299, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.57 }, { "timestamp": 1778139577.886131, "unit": "spark", "load1": 1.86, "load5": 2.19, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102306, "mem_free_mb": 10065, "mem_available_mb": 22304, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139581.357099, "unit": "spark", "load1": 1.95, "load5": 2.21, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102304, "mem_free_mb": 10067, "mem_available_mb": 22306, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.63 }, { "timestamp": 1778139584.8289878, "unit": "spark", "load1": 1.87, "load5": 2.19, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102306, "mem_free_mb": 10065, "mem_available_mb": 22304, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.53 }, { "timestamp": 1778139588.323694, "unit": "spark", "load1": 1.8, "load5": 2.17, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102299, "mem_free_mb": 10071, "mem_available_mb": 22311, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.61 }, { "timestamp": 1778139591.804739, "unit": "spark", "load1": 1.8, "load5": 2.17, "load15": 1.6, "mem_total_mb": 124610, "mem_used_mb": 102300, "mem_free_mb": 10069, "mem_available_mb": 22310, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 46.24 }, { "timestamp": 1778139595.249085, "unit": "spark", "load1": 1.74, "load5": 2.15, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102297, "mem_free_mb": 10071, "mem_available_mb": 22313, "gpu_temp_c": 66.0, "gpu_util_pct": 96.0, "gpu_power_w": 45.93 }, { "timestamp": 1778139598.7523232, "unit": "spark", "load1": 1.68, "load5": 2.13, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102290, "mem_free_mb": 10078, "mem_available_mb": 22320, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 33.68 }, { "timestamp": 1778139599.075281, "unit": "spark", "load1": 1.68, "load5": 2.13, "load15": 1.59, "mem_total_mb": 124610, "mem_used_mb": 102287, "mem_free_mb": 10081, "mem_available_mb": 22323, "gpu_temp_c": 61.0, "gpu_util_pct": 0.0, "gpu_power_w": 23.69 } ] } ], "baseline_p50_s": 3.3814984375203494 } }