DAIBench (DiDi AI Benchmarks
) aims to provide a set of AI evaluation sets for production environments, spanning different types of GPU servers and cloud environments, to provide users with effective and credible test results for future hardware selection , software and library optimization, business model improvement, link stress testing and other stages to lay a solid data foundation and technical reference.
DAIBench comprehensively considers the existing GPU performance testing tools, and divides the indicators into hardware layer, framework (operator) layer, and algorithm layer.
For each level, DAIBench currently supports the following tests:
Layer | Supported Test |
---|---|
Hardware layer | Focusing on the indicators of the hardware itself, such as peak computing throughput (TFLOPS/TOPS) calculation indicators and memory access bandwidth, PCIe communication bandwidth and other I/O indicators. |
Frame/operator layer | Evaluating the computing power of commonly used operators (convolution, Softmax, matrix multiplication, etc.) based on mainstream AI frameworks. |
Model layer | Performing end-to-end evaluation by selecting models in a series of production tasks. |
cd <test_folder>
bash install.sh
bash run.sh
For GPU test, please install suitable nvidia-driver
and cuda
first.
Current operator layer is using DeepBench
cd operator
bash install.sh # download source code & prepare nccl
To run GEMM, convolution, recurrent op and sparse GEMM benchmarks:
bin/gemm_bench <inference|train> <int8|float|half>
To execute the NCCL single All-Reduce benchmark:
bin/nccl_single_all_reduce <num_gpus>
The NCCL MPI All-Reduce benchmark can be run using mpirun as shown below:
mpirun -np <num_ranks> bin/nccl_mpi_all_reduce
num_ranks cannot be greater than the number of GPUs in the system.
docker
and nvidia-docker
is required for model testing. To run specific model, please read Readme.md
in the folder.
General test procedure:
See wiki
for guidelines.
Welcome to contribute by creating issues or sending pull requests. See Contributing Guide
for guidelines.
DAIBench is licensed under the Apache License 2.0
.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。