Friday, August 14, 2015

Hardware for Deep Learning


http://devblogs.nvidia.com/parallelforall/easy-multi-gpu-deep-learning-digits-2/

Figure 3 shows how using multiple GPUs can reduce training time. The graph plots the speedup for training GoogleNet on 1, 2 and 4 GPUS with a batch size of 128. These results were obtained with a DIGITS DevBox using GeForce TITAN X GPUs and the Caffe framework.







Training Speedup Achieved with DIGITS on Multiple GeForce TITAN X GPUs in a DIGITS DevBox. These results were obtained with the Caffe framework and a batch size of 128.


http://arxiv.org/pdf/1404.5997v2.pdf



https://www.linkedin.com/pulse/20141013042457-89310056-which-gpu-to-use-for-deep-learning
https://timdettmers.wordpress.com/2015/03/09/deep-learning-hardware-guide/
https://timdettmers.wordpress.com/2014/09/21/how-to-build-and-use-a-multi-gpu-system-for-deep-learning/
https://timdettmers.wordpress.com/2014/08/14/which-gpu-for-deep-learning/
https://www.quora.com/I-want-to-build-a-workstation-for-deep-learning-practice-What-are-some-suggestions-of-what-to-buy
https://www.quora.com/Would-you-build-a-multi-GPU-system-for-deep-learning-with-GTX-Titan-X-or-Tesla-K40-K80-What-are-the-pros-and-cons
http://pjreddie.com/darknet/hardware-guide/
https://www.facebook.com/notes/chris-lengerich/build-your-own-nvidia-devbox/10152999419281541


Specialized system
https://developer.nvidia.com/devbox
The DIGITS™ DevBox Access Program is available for purchase to qualified deep learning researchers in the United States and will be priced at $15,000. Lead time is 8-10 weeks from payment confirmation.

Amazon EC2 - GPU G2:
https://aws.amazon.com/ec2/instance-types/#g2
https://aws.amazon.com/ec2/pricing/


GPU Cloud Computing
http://www.nvidia.com/object/gpu-cloud-computing-services.html


RAM
http://www.tomshardware.com/forum/299012-30-latency-speed
http://www.crucial.com/usa/en/memory-performance-speed-latency


1 comment:

  1. May 25, 2016
    https://plus.google.com/u/0/112296997936295389954/posts/EYhEggQ2PxE?cfem=1

    ReplyDelete