DeepBench Inference: RNN and Sparse GEMM

Rounding out the last of our DeepBench inference tests are RNN and Sparse GEMM, both available in single precision only. That being said, the FP16 parameter could be selected anyway. Given the low results all around, this is more of an artifact than anything else.

DL Inference: DeepBench - RNN (LSTM)

DL Inference: DeepBench - RNN (GRU)

DL Inference: DeepBench - Sparse GEMM

While RNNs might also be accelerated, DeepBench and NVIDIA only support single precision RNN inference at this time.

DeepBench Inference: Convolutions NVIDIA Caffe2 Docker: ResNet50 and ImageNet
POST A COMMENT

64 Comments

View All Comments

  • Ryan Smith - Tuesday, July 03, 2018 - link

    To clarify: SXM3 is the name of the socket used for the mezzanine form factor cards for servers. All Titan Vs are PCie. Reply
  • Drumsticks - Tuesday, July 03, 2018 - link

    Nice review. Will anandtech be putting forth an effort to cover the ML hardware space in the future? AMD and Intel both seem to have plans here.

    The V100 and Titan V should have well over 100TF according to Nvidia in training and inference, if I remember correctly, but nothing I saw here got close in actuality. Were these benches not designed to hit those numbers, or are those numbers just too optimistic in most scenarios to occur?
    Reply
  • Ryan Smith - Tuesday, July 03, 2018 - link

    "The V100 and Titan V should have well over 100TF according to Nvidia in training and inference"

    The Titan V only has 75% of the memory bandwidth of the V100. So it's really hard to hit 100TF. Even in our Titan V preview where we ran a pure CUDA-based GEMM benchmark, we only hit 97 TFLOPS. Meanwhile real-world use cases are going to be lower still, as you can only achieve those kinds of high numbers in pure tensor core compute workloads.

    https://www.anandtech.com/show/12170/nvidia-titan-...
    Reply
  • Nate Oh - Tuesday, July 03, 2018 - link

    To add on to Ryan's comment, 100+ TF is best-case (i.e. synthetic) performance based on peak FMA ops on individual matrix elements, which only comes about when everything perfectly qualifies for tensor core acceleration, no memory bottleneck by reusing tons of register data, etc. Reply
  • remedo - Tuesday, July 03, 2018 - link

    Nate, I hope you could have included more TensorFlow/Keras specific benchmarks, given that the majority of deep learning researchers/developers are now using TensorFlow. Just compare the GitHub stats of TensorFlow vs. other frameworks. Therefore, I feel that this article missed some critical benchmarks in that regard. Still, this is a fascinating article, and thank you for your work. I understand that Anandtech is still new to deep learning benchmarks compared to your decades of experience in CPU/Gaming benchmark. If possible, please do a future update! Reply
  • Nate Oh - Tuesday, July 03, 2018 - link

    Several TensorFlow benchmarks did not make the cut for today :) We were very much interested in using it, because amongst other things it offers global environmental variables to govern tensor core math, and integrates somewhat directly with TensorRT. However, we've been having issues finding and using one that does all the things we need it to do (and also offers different results than just pure throughput), and I've gone so far as trying to rebuild various models/implementations directly in Python (obviously to no avail, as I am ultimately not an ML developer).

    According to people smarter than me (i.e. Chintala, and I'm sure many others), if it's only utilizing standard cuDNN operations then frameworks should perform about the same; if there are significant differences, a la the inaugural version of Deep Learning Frameworks Comparison, it is because it is poorly optimized for TensorFlow or whatever given framework. From a purely GPU performance perspective, usage of different frameworks often comes down to framework-specific optimization, and not all reference implementations or benchmark suite tests do what we need it to do out-of-the-box (not to mention third-party implementations). Analyzing the level of TF optimization is developer-level work, and that's beyond the scope of the article. But once benchmark suites hit their stride, that will resolve that issue for us.

    For Keras, I wasn't able to find anything that was reasonably usable by a non-developer, though I could've easily missed something (I'm aware of how it relates to TF, Theano, MXNet, etc). I'm sure that if we replaced PyTorch with Tensorflow implementations, we would get questions on 'Where's PyTorch?' :)

    Not to say your point isn't valid, it is :) We're going to keep on looking into it, rest assured.
    Reply
  • SirPerro - Thursday, July 05, 2018 - link

    Keras has some nice examples in its github repo to be run with the tensorflow backend but for the sake of benchmarking it does not offer anything that it's not covered by the pure tensorflow examples, I guess Reply
  • BurntMyBacon - Tuesday, July 03, 2018 - link

    I believe the GTX Titan with memory clock 6Gbps and memory bus width of 384 bits should have a memory bandwidth of 288GB/sec rather than the list 228GB/sec. Putting that aside, this is a nice review. Reply
  • Nate Oh - Tuesday, July 03, 2018 - link

    Thanks, fixed Reply
  • Jon Tseng - Tuesday, July 03, 2018 - link

    Don't be silly. All we care about is whether it can run Crysis at 8K. Reply

Log in

Don't have an account? Sign up now