NVIDIA DGX-1 Community-Based Benchmark
1 like1,109 views
The document discusses a community-based deep learning benchmark using an NVIDIA DGX-1 supercomputer. It announces that the benchmark will be set up by mid-March 2017 and interested participants should contact them. It also summarizes previous benchmarks conducted on different GPUs and frameworks, comparing efficiency when training various neural networks. Details are provided on benchmarks measuring minibatch efficiency for TensorFlow. Participants are directed to a blog post for more information.
NVIDIA DGX-1 Community-Based Benchmark
- 1. NVIDIA DGX-1 Supercomputer Community-Based Deep Learning Benchmark
- 2. Community-Based Deep Learning Benchmark We have an NVIDIA DGX-1 available for testing and would like to organise a Community-Driven Benchmark We plan to have the benchmark set-up by mid-March 2017. If you are interested in running your tests, contact us asap Please check our Blogpost for info: bit.ly/DGX1-Benchmark IMPORTANT:
- 3. Benchmark we’ve already done on NVIDIA GTX1080 -TITAN X Maxwell -TITAN X Pascal - K40 PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 4. GPU vs Framework vs Network - K40 This Benchmark compares the efficiency of various GPUs when training different networks using different frameworks. This is a common approach that allows us to decide on which framework to use for a given GPU and a given architecture PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 5. GPU vs Framework vs Network -TITAN X Maxwell This Benchmark compares the efficiency of various GPUs when training different networks using different frameworks. This is a common approach that allows us to decide on which framework to use for a given GPU and a given architecture PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 6. GPU vs Framework vs Network - GTX 1080 This Benchmark compares the efficiency of various GPUs when training different networks using different frameworks. This is a common approach that allows us to decide on which framework to use for a given GPU and a given architecture PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 7. GPU vs Framework vs Network -TITAN X Pascal This Benchmark compares the efficiency of various GPUs when training different networks using different frameworks. This is a common approach that allows us to decide on which framework to use for a given GPU and a given architecture PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 8. Minibatch Efficiency forTensorFlow This Benchmark compares the mini-batch efficiency for each GPU when training different architectures. As a rule of thumb, larger minibatch size means more efficient training, however be prepare for a few surprise PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 9. Minibatch Efficiency forTensorFlow This Benchmark compares the mini-batch efficiency for each GPU when training different architectures. As a rule of thumb, larger minibatch size means more efficient training, however be prepare for a few surprise PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 10. Minibatch Efficiency forTensorFlow This Benchmark compares the mini-batch efficiency for each GPU when training different architectures. As a rule of thumb, larger minibatch size means more efficient training, however be prepare for a few surprise PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 11. Minibatch Efficiency forTensorFlow This Benchmark compares the mini-batch efficiency for each GPU when training different architectures. As a rule of thumb, larger minibatch size means more efficient training, however be prepare for a few surprise PleasecheckourBlogpostforinfo:bit.ly/DGX1-Benchmark
- 12. it.linkedin.com/in/ebusto Contact me for any other question