Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by reitzensteinm on Sunday October 29, 2017 @07:13PM
(#15582086)
It's probably better to analogize it to Xeon, which prints Intel roughly as much cash as the tax revenues of Croatia.
There are plenty of things to arbitrarily segment in a GPU - HBM, large memory sizes, tensor cores, FP64.
The problem with hard disk segmentation wasn't the idea of segmenting at all, it's that there were no good ways of doing it. A higher MTBF? You're better off buying more, cheaper disks. Density (eg from helium)? Might be worth a 20-50% premium in $/gb, but not 200-500% $/transistor as with Xeon.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by microcolonel on Sunday October 29, 2017 @05:54PM
(#15581644)
There are some other knobs that NVIDIA will surely try (has surely tried?) to tweak as well, such as making the consumer devices deliberately less power efficient, fusing off chip features which make compute workloads more efficient, or sabotaging the drivers (either in general or on a per-application basis).
With disks there was always just as much pressure on the consumer side to keep energy down, cost down, and capacity up, which means that there was no natural segmentation, and no straightforward unnatural segmentation.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by llukas on Monday October 30, 2017 @12:40AM
(#15583134)
I think you're vastly underestimating amount of work that goes into those things. What if instead what you suggest they do not invest as hard in gaming parts? Design, QA, QC and support cost money after all.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by Jabbles on Sunday October 29, 2017 @05:57PM
(#15581655)
Is the initial cost of purchase the largest factor? Or are running costs enough to tip the balance?
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by jamesblonde on Sunday October 29, 2017 @06:05PM
(#15581713)
Read the article. Running costs can be quite high (up to 1000 dollars/months with colocation/energy), but they are still massively cheaper than running 24x7 the same GPUs in the cloud.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by Bromskloss on Sunday October 29, 2017 @06:49PM
(#15581976)
I took the question to be about the choice between the two products.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by shaklee3 on Sunday October 29, 2017 @06:43PM
(#15581944)
Where did you get that the 1080ti is 60% of the performance of the p100? It is slightly higher performance than the p100, but rated with a lower mtbf. Of course, you end up with many crippled features, like RDMA and dynamic parallelism, which most people probably don't need anyways.
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by jamesblonde on Monday October 30, 2017 @12:40AM
(#15583132)
Deep learning training is basically 2 phases: the forward pass and the backwards pass (backpropagation). There are no loops. As such, most DL training is memory bound.
Memory bandwidth of 1080TI is 484.4 GB/s
Memory bandwidth of p100 is 720.9 GB/s
Memory bandwidth of v100 is 900.1 GB/s
Re:DeepLearning11: 10x Nvidia GTX 1080 Ti Single Root Deep L...
by shaklee3 on Monday October 30, 2017 @09:28AM
(#15584823)
For applications that are not memory bound, the 1080ti is better. Also, if we are talking about deep learning, inference can be handled with lower precision. The p100 does not have int8 support and hardware, but the 1080ti does. So presumably for the same memory bandwidth you get twice as much data.
The 1080Ti card is about 60% of the performance of the P100, but costs 700 dollars instead of 5k dollars. Of course, people will try and build these boxes, they have a much higher ROI compared to Nvidia's DGX-1. So what do NVidia do? Try and stop vendors from selling them! https://www.pcgamesn.com/nvidia-geforce-server