# Example Benchmark Output This document shows example output from the benchmark suite. ## Quick Benchmark Output ``` ============================================================ Quick Performance Benchmark ============================================================ Running subset of benchmarks for quick verification... For full benchmarks, use: python run_all_benchmarks.py ============================================================ Running benchmark: Quick Voxel Ray Casting ============================================================ Warmup (3 iterations)... Running (20 iterations)... Progress: 10/20 Progress: 20/20 ============================================================ Results for: Quick Voxel Ray Casting ============================================================ Duration: 458.23 ms Throughput: 43.65 FPS Latency (p50): 21.84 ms Latency (p95): 25.32 ms Latency (p99): 27.15 ms CPU Util: 42.3% Memory: 987.45 MB GPU Util: 0.0% GPU Memory: 0.00 MB No performance regressions detected. ============================================================ Running benchmark: Quick Voxel Updates ============================================================ Warmup (3 iterations)... Running (30 iterations)... Progress: 10/30 Progress: 20/30 Progress: 30/30 ============================================================ Results for: Quick Voxel Updates ============================================================ Duration: 234.56 ms Throughput: 127.94 FPS Latency (p50): 7.45 ms Latency (p95): 8.92 ms Latency (p99): 9.34 ms CPU Util: 38.7% Memory: 856.23 MB GPU Util: 0.0% GPU Memory: 0.00 MB No performance regressions detected. Saved results to benchmark_results/results_20251113_143022.json Saved CSV to benchmark_results/results_20251113_143022.csv ============================================================ Quick Benchmark Complete ============================================================ Results saved to: benchmark_results/ ``` ## Main Benchmark Suite Output ``` ============================================================ PixelToVoxel Performance Benchmark Suite ============================================================ ============================================================ Running benchmark: Voxel Ray Casting (500^3) ============================================================ Warmup (5 iterations)... Running (50 iterations)... Progress: 10/50 Progress: 20/50 Progress: 30/50 Progress: 40/50 Progress: 50/50 ============================================================ Results for: Voxel Ray Casting (500^3) ============================================================ Duration: 1234.56 ms Throughput: 40.51 FPS Latency (p50): 23.45 ms Latency (p95): 27.89 ms Latency (p99): 30.12 ms CPU Util: 45.2% Memory: 2134.56 MB GPU Util: 0.0% GPU Memory: 0.00 MB No performance regressions detected. ============================================================ Running benchmark: Motion Detection (8K) ============================================================ Warmup (5 iterations)... Running (50 iterations)... Progress: 10/50 Progress: 20/50 Progress: 30/50 Progress: 40/50 Progress: 50/50 ============================================================ Results for: Motion Detection (8K) ============================================================ Duration: 2345.67 ms Throughput: 21.32 FPS Latency (p50): 45.23 ms Latency (p95): 48.76 ms Latency (p99): 51.34 ms CPU Util: 67.8% Memory: 2567.89 MB GPU Util: 0.0% GPU Memory: 0.00 MB No performance regressions detected. Saved results to benchmark_results/results_20251113_143530.json Saved CSV to benchmark_results/results_20251113_143530.csv Generated report: benchmark_results/report_20251113_143530.html ============================================================ Save these results as performance baseline? (y/n): y Saved 3 baselines to benchmark_results/baselines.json ============================================================ Benchmark suite completed! ============================================================ ``` ## Camera Benchmark Output ``` ============================================================ Benchmarking 8K Video Decode Performance ============================================================ Generating synthetic 8K frames (7680x4320)... Processed 50/300 frames Processed 100/300 frames Processed 150/300 frames Processed 200/300 frames Processed 250/300 frames Processed 300/300 frames Results: Avg Decode Time: 42.35 ms Decode FPS: 23.61 Max FPS: 28.45 p95 Latency: 45.67 ms p99 Latency: 48.92 ms ============================================================ Benchmarking Motion Extraction Throughput ============================================================ Testing 8K (7680x4320)... Processed 50/300 frames Processed 100/300 frames Processed 150/300 frames Processed 200/300 frames Processed 250/300 frames Processed 300/300 frames Avg Motion Time: 38.45 ms Motion FPS: 26.01 p99 Latency: 42.34 ms Testing 4K (3840x2160)... Processed 50/300 frames Processed 100/300 frames Processed 150/300 frames Processed 200/300 frames Processed 250/300 frames Processed 300/300 frames Avg Motion Time: 9.23 ms Motion FPS: 108.35 p99 Latency: 11.45 ms Testing 1080p (1920x1080)... Processed 50/300 frames Processed 100/300 frames Processed 150/300 frames Processed 200/300 frames Processed 250/300 frames Processed 300/300 frames Avg Motion Time: 2.34 ms Motion FPS: 427.35 p99 Latency: 3.12 ms ============================================================ Results saved to: benchmark_results/camera/camera_benchmark_20251113_144022.json ============================================================ ``` ## CUDA Benchmark Output ``` ======================================== CUDA Voxel Benchmark Suite ======================================== GPU: NVIDIA GeForce RTX 3080 Compute Capability: 8.6 Global Memory: 10.00 GB Multiprocessors: 68 Max Threads/Block: 1024 Benchmarking Ray Casting (500^3 grid, 100000 rays)... ======================================== Benchmark: Voxel Ray Casting (DDA) ======================================== Duration: 8.45 ms Throughput: 68.34 GOPS Memory BW: 234.56 GB/s Kernel Time: 8.23 ms Blocks: 391 Threads/Block: 256 ======================================== Benchmarking Voxel Updates (500^3 grid, 1000000 updates)... ======================================== Benchmark: Voxel Updates (Atomic) ======================================== Duration: 4.23 ms Throughput: 236.41 GOPS Memory BW: 12.34 GB/s Kernel Time: 4.12 ms Blocks: 3907 Threads/Block: 256 ======================================== Benchmarking Memory Bandwidth (125000000 elements)... ======================================== Benchmark: Memory Bandwidth (Coalesced) ======================================== Duration: 1.23 ms Throughput: 101.63 GOPS Memory BW: 406.50 GB/s Kernel Time: 1.23 ms Blocks: 488282 Threads/Block: 256 ======================================== Benchmark suite completed! ``` ## Network Benchmark Output ``` ============================================================ Benchmarking TCP Throughput (10s) ============================================================ Results: Bytes Sent: 8,456,789,012 Duration: 10.02 s Throughput: 6,749.23 Mbps ============================================================ Benchmarking UDP Throughput (10s) ============================================================ Results: Packets Sent: 7,142,857 Packets Received: 7,135,324 Packet Loss: 0.11% Throughput: 7,999.45 Mbps ============================================================ Benchmarking TCP Latency (1000 pings) ============================================================ Progress: 100/1000 Progress: 200/1000 Progress: 300/1000 Progress: 400/1000 Progress: 500/1000 Progress: 600/1000 Progress: 700/1000 Progress: 800/1000 Progress: 900/1000 Progress: 1000/1000 Results: Avg Latency: 0.23 ms p50 Latency: 0.21 ms p95 Latency: 0.34 ms p99 Latency: 0.45 ms ============================================================ Benchmarking Multi-Client Scalability (10 clients) ============================================================ Results: Clients Completed: 10/10 Total Bytes: 12,345,678,901 Aggregate Throughput: 9,876.54 Mbps Per-Client Avg: 987.65 Mbps ============================================================ Results saved to: benchmark_results/network/network_benchmark_20251113_144530.json ============================================================ ``` ## Full Suite Output ``` ====================================================================== PixelToVoxel Comprehensive Benchmark Suite ====================================================================== Started: 2025-11-13 14:45:30 Checking environment... Python: 3.11.14 numpy: OK cv2: OK matplotlib: OK psutil: OK CUDA: OK ====================================================================== Running Main Benchmark Suite ====================================================================== [... main suite output ...] ✓ Main benchmark suite completed ====================================================================== Running Camera Benchmark Suite ====================================================================== [... camera suite output ...] ✓ Camera benchmark suite completed ====================================================================== Running CUDA Voxel Benchmarks ====================================================================== [... CUDA output ...] ✓ CUDA benchmark suite completed ====================================================================== Running Network Benchmark Suite ====================================================================== [... network output ...] ✓ Network benchmark suite completed Combined results saved to: benchmark_results/combined_results_20251113_144530.json Summary saved to: benchmark_results/summary_20251113_144530.txt ====================================================================== Benchmark Suite Completed ====================================================================== Total Duration: 487.3 seconds Results saved to: /home/user/Pixeltovoxelprojector/tests/benchmarks/benchmark_results ``` ## HTML Report Example The HTML report includes: ### Summary Section ``` Avg Throughput: 45.2 FPS Avg Latency: 24.3 ms Avg CPU Usage: 52% Avg GPU Usage: 78% ``` ### Performance Charts - Throughput Comparison (bar chart) - Latency Distribution (grouped bar chart with p50/p95/p99) - Resource Utilization (CPU/GPU utilization and memory) ### Detailed Results Table | Benchmark | Throughput (FPS) | p50 (ms) | p95 (ms) | p99 (ms) | CPU % | GPU % | Memory (MB) | Status | |-----------|------------------|----------|----------|----------|-------|-------|-------------|--------| | Voxel Ray Casting | 40.51 | 23.45 | 27.89 | 30.12 | 45.2 | 0.0 | 2134.6 | PASS | | Motion Detection | 21.32 | 45.23 | 48.76 | 51.34 | 67.8 | 0.0 | 2567.9 | PASS | | Voxel Updates | 127.94 | 7.45 | 8.92 | 9.34 | 38.7 | 0.0 | 856.2 | PASS | ## CSV Output Example ```csv name,duration_ms,throughput_fps,latency_p50_ms,latency_p95_ms,latency_p99_ms,cpu_percent,memory_mb,gpu_percent,gpu_memory_mb,timestamp Voxel Ray Casting (500^3),1234.56,40.51,23.45,27.89,30.12,45.2,2134.56,0.0,0.00,2025-11-13T14:35:30.123456 Motion Detection (8K),2345.67,21.32,45.23,48.76,51.34,67.8,2567.89,0.0,0.00,2025-11-13T14:37:15.654321 Voxel Grid Updates,187.23,127.94,7.45,8.92,9.34,38.7,856.23,0.0,0.00,2025-11-13T14:37:22.987654 ``` ## Performance Regression Example ``` ============================================================ Results for: Voxel Ray Casting (500^3) ============================================================ Duration: 1456.78 ms Throughput: 34.32 FPS Latency (p50): 27.89 ms Latency (p95): 32.45 ms Latency (p99): 35.67 ms CPU Util: 48.9% Memory: 2345.67 MB GPU Util: 0.0% GPU Memory: 0.00 MB WARNING: Performance regressions detected: - Throughput regression: 34.32 < 36.45 FPS - Latency regression: 35.67 > 30.12 ms ``` ## Summary Text File Example ``` ====================================================================== PixelToVoxel Benchmark Summary ====================================================================== Started: 2025-11-13 14:45:30 Finished: 2025-11-13 14:53:37 Duration: 487.3 seconds ====================================================================== Suite Status ====================================================================== main_suite ✓ PASS camera_suite ✓ PASS cuda_suite ✓ PASS network_suite ✓ PASS ====================================================================== Main Suite Results ====================================================================== Voxel Ray Casting (500^3) Throughput: 40.51 FPS p99 Latency: 30.12 ms Motion Detection (8K) Throughput: 21.32 FPS p99 Latency: 51.34 ms Voxel Grid Updates Throughput: 127.94 FPS p99 Latency: 9.34 ms ====================================================================== End of Summary ====================================================================== ```