...

https://github-dev.cs.illinois.edu/kindrtnk/DL

Official TF Benchmark System Statistics

• Instance type: NVIDIA® DGX-1™
• GPU: 8x NVIDIA® Tesla® P100
• OS: Ubuntu 16.04 LTS with tests run via Docker
• CUDA / cuDNN: 8.0 / 5.1
• TensorFlow GitHub hash: b1e174e
• Benchmark GitHub hash: 9165a70
• Build Command:bazel build -c opt --copt=-march="haswell" --config=cuda //tensorflow/tools/pip_package:build_pip_package
• Disk: Local SSD
• DataSet: ImageNet
• Test Date: May 2017

System Statistics (more details in GitHub Repo)

* Instance type: IBM Power9 Hal000, 8335-GTG AC922 server
* CPU: 2x 20-core IBM POWER9 CPU @ 2.00GHz
* SDRAM: 512G DDR4
* GPU: 4x NVIDIA® Tesla® V100, 5120 cores, 16 GB HBM 2
* OS: Red Hat Enterprise Linux Server release 7.4
* Python Distribution: Anaconda python 3.6.2
* CUDA / cuDNN: 9.1/7.0.5
* TensorFLow Version: 1.5.0
* Disk: Local SSD
* DataSet: ImageNet (synthetic)
* Precision: floating point 32 and 16
* Test Date: Mar 25 2018POWER9 (hal000)

...