GeoWave3D
AI 科研综合平台
GeoWave3D/main main
CUDA 13.1 · NVIDIA L20
src/kernel.cu AI 建议

Experiment 014

Completed Testing Agent

8th-order Acoustic Simulation

Configuration
Backend
CUDA 13.1
Precision
FP32
Grid
951 × 501 × 510
Grid Spacing
10 m
Time Step
1 ms
Steps
1500
Execution
Agent
Testing Agent
GPU
NVIDIA L20 · 89%
Code
a83c12f
Started
14:02:14
End
14:02:27
Results
Total Runtime
12.84 s
Kernel Runtime
6.48 s
Kernel Step
4.32 ms
Memory
8.7 GB
Max Error
2.1e-3
Throughput
89.4 GB/s
Artifacts
Data wavefield_001.npy Benchmark benchmark.json Log run.log Figure wavefield_t1.5s.svg

Wavefield · t = 1.5 s

Completed Visualization Agent

z-slice 255 · 8 阶声波快照

t = 1.5 s · z-slice 255 0250 500750950 5004003002001000 x (m) z (m) +1.00−1.0
Metadata
Grid951×501×510
Time1.5 s
ColormapSeismic
Range−1.0 / +1.0
Grid Spacing10 m
Provenance
Generated ByVisualization Agent
ModelGemini
SkillWavefield Plot
Inputwavefield_001.npy
ExperimentExperiment 014
Codea83c12f
AI Analysis
波前呈椭圆近似理想;x 方向采样在远场略低于 Nyquist,建议网格加密或插值后再发布。

Benchmark L20

Completed Performance Agent

NVIDIA L20 · 1500 Steps

目标 4.5 ms 基线共享内存向量化LaunchBnd常量内存CUDA Graph 5.824.313.64 3.123.052.94 ms / step
配置ms/stepGB/s加速比
基线(全局读 ×24)5.8239.11.00×
共享内存 tile4.3152.81.35×
向量化 float43.6462.51.60×
__launch_bounds__(128)3.1272.91.87×
常量内存 FD_W3.0574.61.91×
CUDA Graph 捕获2.9477.41.98×
Source Experiment 014 · Artifact perf_scaling_L20.json · Code a83c12f

Code Analysis · src/kernel.cu

Completed Code Analysis Agent

由 Code Analysis Agent 以 GPT-5.6 分析 · 产出 analysis/findings.json

Overview
WARN2 Warnings · 1 Hotspot  ·  src/kernel.cu · 33 行 · 8 阶声波 stencil · FP32
Structure
INFO单内核 stencil 设计,__constant__ 权重 + __restrict__ 指针提示无别名。
INFOleapfrog 时间步进在 L31solver.cpp 的 swap 保持一致。
Hotspots
HOTkernel.cu:26 热点循环 · 8 阶模板每网格点 24 次全局读 · 建议共享内存 tile 削减访存。
Risks
WARNPotential uncoalesced memory access · kernel.cu:26 · Medium
WARNRegister pressure from 8th-order expansion · kernel.cu:25 · Medium
Correctness
OKhalo guard 与 8 阶模板索引范围一致;跳界处理正确。
Dependencies
DEPkernel.cusolver.cpp · wavefields.h · __constant__ FD_W
DEPLaunched from solver.cpp:21 · run_experiment 内核启动
main src/kernel.cu Ln 26, Col 1UTF-8LFC++ / CUDA空格: 2