# Quark Profiling Results memory_usage: - step: "Start" timestamp: 1775811499.6711347 relative_time_secs: 0.0 cpu_memory_mb: 3963.88 gpu_memory_mb: 783.08 disk_read_mb: 0.0 disk_write_mb: 0.0 - step: "Model Loading Start" timestamp: 1775811499.7619655 relative_time_secs: 0.09083080291748047 cpu_memory_mb: 3963.88 gpu_memory_mb: 783.08 disk_read_mb: 0.0 disk_write_mb: 0.0 - step: "Model Loading End" timestamp: 1775811532.1563113 relative_time_secs: 32.48517656326294 cpu_memory_mb: 4173.82 gpu_memory_mb: 60728.46 disk_read_mb: 59649.02 disk_write_mb: 2.06 - step: "Dataset Loading Start" timestamp: 1775811551.9537203 relative_time_secs: 52.28258562088013 cpu_memory_mb: 9057.3 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Dataset Loading End" timestamp: 1775811555.4571419 relative_time_secs: 55.78600716590881 cpu_memory_mb: 9059.46 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Model Quantization Start" timestamp: 1775811555.5480177 relative_time_secs: 55.876883029937744 cpu_memory_mb: 9059.46 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Model Preparation Start" timestamp: 1775811555.6379032 relative_time_secs: 55.96676850318909 cpu_memory_mb: 9059.46 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Model Preparation End" timestamp: 1775811555.7979052 relative_time_secs: 56.12677049636841 cpu_memory_mb: 9059.47 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Advanced Algorithms Start" timestamp: 1775811555.887473 relative_time_secs: 56.21633839607239 cpu_memory_mb: 9059.47 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Advanced Algorithms End" timestamp: 1775811555.976544 relative_time_secs: 56.30540919303894 cpu_memory_mb: 9059.47 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Calibration Start" timestamp: 1775811556.0654619 relative_time_secs: 56.39432716369629 cpu_memory_mb: 9059.47 gpu_memory_mb: 61855.47 disk_read_mb: 59679.74 disk_write_mb: 1030.27 - step: "Calibration End" timestamp: 1775811556.8114972 relative_time_secs: 57.14036250114441 cpu_memory_mb: 9086.19 gpu_memory_mb: 118389.49 disk_read_mb: 59679.74 disk_write_mb: 1058.97 - step: "Model Quantization End" timestamp: 1775811556.900997 relative_time_secs: 57.229862213134766 cpu_memory_mb: 9086.19 gpu_memory_mb: 115505.49 disk_read_mb: 59679.74 disk_write_mb: 1058.97 - step: "Freeze Model Start" timestamp: 1775811556.9907522 relative_time_secs: 57.31961750984192 cpu_memory_mb: 9086.19 gpu_memory_mb: 112649.49 disk_read_mb: 59679.74 disk_write_mb: 1058.97 - step: "Freeze Model End" timestamp: 1775811558.4109159 relative_time_secs: 58.73978114128113 cpu_memory_mb: 9742.0 gpu_memory_mb: 72820.82 disk_read_mb: 59688.81 disk_write_mb: 1058.98 - step: "Export HF Safetensors Start" timestamp: 1775811558.5011773 relative_time_secs: 58.830042600631714 cpu_memory_mb: 9743.67 gpu_memory_mb: 72820.84 disk_read_mb: 59688.81 disk_write_mb: 1058.98 - step: "Export HF Safetensors End" timestamp: 1775811576.8544993 relative_time_secs: 77.18336462974548 cpu_memory_mb: 24957.37 gpu_memory_mb: 71846.62 disk_read_mb: 59688.81 disk_write_mb: 19712.72 - step: "End" timestamp: 1775811577.0333073 relative_time_secs: 77.36217260360718 cpu_memory_mb: 24882.35 gpu_memory_mb: 71846.62 disk_read_mb: 59688.81 disk_write_mb: 19712.72 # Summary Metrics total_quantization_time_seconds: 77.3622 peak_memory_mb: 63784.82 peak_gpu_memory_mb: 118389.49 total_disk_read_mb: 59688.81 total_disk_write_mb: 19712.72 # Metric Definitions: # # Checkpoint Metrics (per record): # - step: Name of the profiling checkpoint. Common steps include: # - "Start": Initial state when profiling begins # - "Model Loaded": After loading the ONNX model into memory # - "Pre-process Start/End": Before and after model preprocessing # - "Calibration Start/End": Before and after calibration data collection # - "Quantization (MatMulNBits) Start/End": MatMulNBits quantization phase # - "Quantization (Static) Start/End": Static quantization phase # - "Post-process Start/End": Before and after post-processing # - "Fast Finetune Start/End": Before and after fast finetuning (if enabled) # - timestamp: Unix timestamp (seconds since epoch) when this measurement was taken. Useful for correlating with external logs or events. # - relative_time_secs: Time elapsed (in seconds) since the "Start" step. Useful for understanding the duration of each phase relative to the beginning of profiling. # - cpu_memory_mb: Current Resident Set Size (RSS) in megabytes at this step. This includes memory from the main process and all child processes. RSS represents the portion of memory held in RAM (not swapped out). # - gpu_memory_mb: Current GPU memory usage in megabytes. This represents actual GPU memory used by the process, including allocations from PyTorch, ONNX Runtime, TensorRT, and other frameworks. Only available when PyTorch with CUDA/ROCm is installed and GPU is available. # - disk_read_mb: Cumulative disk bytes read (in megabytes) since the start of profiling. Measured relative to the baseline captured at the 'Start' checkpoint, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # - disk_write_mb: Cumulative disk bytes written (in megabytes) since the start of profiling. Measured relative to the baseline captured at the 'Start' checkpoint, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # # Summary Metrics (overall): # - total_quantization_time_seconds: Total elapsed time (in seconds) from the start of profiling to the end of the quantization process. # - peak_memory_mb: Peak resident set size (RSS) in megabytes for the main process during the entire profiling session. On Linux, this is read from VmHWM (high water mark) in /proc//status. On Windows, this is the peak working set size. This metric may not be available on all platforms. # - peak_gpu_memory_mb: Peak GPU memory usage in megabytes during the entire profiling session. This is the maximum GPU memory used, including allocations from PyTorch, ONNX Runtime, TensorRT, and other frameworks. Only available when PyTorch with CUDA/ROCm is installed and GPU is available. # - total_disk_read_mb: Total disk bytes read (in megabytes) during the entire profiling session. Computed as the difference between the final and baseline cumulative read counters, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # - total_disk_write_mb: Total disk bytes written (in megabytes) during the entire profiling session. Computed as the difference between the final and baseline cumulative write counters, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root).