Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ryzen Threadripper Pro 3995WX Spotted

guru3d.com

Post reply on HN
Print view View original

41–50 of 170 posts

Re: Ryzen Threadripper Pro 3995WX Spotted

#41

Post by CSDude »

vosper wrote: post #40
staticfloat wrote: post #14

Earlier quoted context omitted.

One big difference is memory availability; the amount of RAM directly accessible from a 64-core CPU is much larger than that of the thousands of CUDA cores on a Quadro card. Perhaps for many workloads intelligent use of the PCIE bus can make up for that, streaming datasets in/out of the card, but for others, having random access to hundreds of GBs of data may be non-negotiable. I'd be interested to know what workload…

I’m not sure about “require”, but we use multi-hundred-gigabytes or RAM instances for our Elasticsearch coordinator nodes, which seems to work well.

Elasticsearch recommends <32GB because of 32-bit per their doc, is it different for coordinator nodes?

Re: Ryzen Threadripper Pro 3995WX Spotted

#42

Post by jacquesm »

dTal wrote: post #39
Uehreka wrote: post #12

I feel like 64 cores is getting rather close to a tipping point: What workloads are so massively parallel that they can use 64 cores of x86 but can’t use the thousands of CUDA cores on a Quadro card? I’m sure right now there are workloads that just need particular x86 instructions, but that feels like a temporary problem. Am I wrong about that being a temporary problem (that would feel frustrating)? Are these cores j…

CPUs are much more general! Unscientifically, on my 2950X/1070Ti workstation, Cycles takes approximately the same amount of time to raytrace a scene on the CPU vs the GPU. I think this is a great example of a highly parallel workload that is challenging for GPUs due to lack of coherence - extreme props to the Cycles team for making it work on GPU at all. Also, year-on-year performance increases for nVidia GPUs are re…

The people that use CUDA don't care about the underlying architecture, they need results. Which ever manufacturer will deliver those results will get the $.

Raytracing is not an ideal match workload wise for a GPU, too much divergence, but there are plenty of workloads where GPUs will have an edge for along time to come.

Re: Ryzen Threadripper Pro 3995WX Spotted

#43

Post by jacquesm »

achempion wrote: post #30
imtringued wrote: post #24

Earlier quoted context omitted.

GPUs can't run the same code you would have written on a CPU. If you had the choice between an equally performing CPU and GPU you would always use the CPU.

Why? Isn't the CPU will cost more than GPU because of more general instruction set?

On a per-flops basis yes. But it all depends on your workload, for some workloads GPUs are a better match than CPUs and vice-versa.

Re: Ryzen Threadripper Pro 3995WX Spotted

#44

Post by qayxc »

jacquesm wrote: post #42
dTal wrote: post #39

Earlier quoted context omitted.

CPUs are much more general! Unscientifically, on my 2950X/1070Ti workstation, Cycles takes approximately the same amount of time to raytrace a scene on the CPU vs the GPU. I think this is a great example of a highly parallel workload that is challenging for GPUs due to lack of coherence - extreme props to the Cycles team for making it work on GPU at all. Also, year-on-year performance increases for nVidia GPUs are re…

The people that use CUDA don't care about the underlying architecture, they need results. Which ever manufacturer will deliver those results will get the $. Raytracing is not an ideal match workload wise for a GPU, too much divergence, but there are plenty of workloads where GPUs will have an edge for along time to come.

It's more than that - NVIDIA invested massively in their software ecosystem and mindshare (e.g. by a huge sponsoring program that saw universities flooded with their hardware).

AMD cards offer comparable compute performance and are sometimes even better. What AMD lacks, though, are ready-to-use software packages like CUDA, cuDNN, cuBLAS, and the like.

NVIDIA basically bought their way into the research community and solidified their strong foothold by keeping CUDA proprietary.

It's fascinating how other companies (Microsoft) get bashed on a regular basis for using the exact same tactics (see Direct3D vs OpenGL on Windows), while people are perfectly fine with NVIDIA doing it...

Re: Ryzen Threadripper Pro 3995WX Spotted

#45

Post by sradman »

Uehreka wrote: post #12

I feel like 64 cores is getting rather close to a tipping point: What workloads are so massively parallel that they can use 64 cores of x86 but can’t use the thousands of CUDA cores on a Quadro card? I’m sure right now there are workloads that just need particular x86 instructions, but that feels like a temporary problem. Am I wrong about that being a temporary problem (that would feel frustrating)? Are these cores j…

> What workloads are so massively parallel that they can use 64 cores of x86 but can’t use the thousands of CUDA cores on a Quadro card?

Column stores like SAP HANA. Compressed bitmap indexes leverage SIMD but not floating point matrix math GPUs. The workstation vs server case is narrower though; benchmarking, dev testing, and local MS PowerPivot like analysis on large in-memory datasets.

Re: Ryzen Threadripper Pro 3995WX Spotted

#46

Post by jacquesm »

qayxc wrote: post #44
jacquesm wrote: post #42

Earlier quoted context omitted.

The people that use CUDA don't care about the underlying architecture, they need results. Which ever manufacturer will deliver those results will get the $. Raytracing is not an ideal match workload wise for a GPU, too much divergence, but there are plenty of workloads where GPUs will have an edge for along time to come.

It's more than that - NVIDIA invested massively in their software ecosystem and mindshare (e.g. by a huge sponsoring program that saw universities flooded with their hardware). AMD cards offer comparable compute performance and are sometimes even better. What AMD lacks, though, are ready-to-use software packages like CUDA, cuDNN, cuBLAS, and the like. NVIDIA basically bought their way into the research community and…

Because NVIDIA did not do anything illegal - so far. MS is a convicted monopolist that actively went after competitors rather than just to let the market decide. What NVIDIA has done so far is perfectly fine in my book, paying companies not to use competitor hardware, funding third party lawsuits and stealing IP, the browser wars and so on are things that Microsoft has done.

Re: Ryzen Threadripper Pro 3995WX Spotted

#47

Post by tinco »

fomine3 wrote: post #6

> The big news is that the Asia website reported the 3995WX is an offering 8-channel DDR4 memory interface, up to 2 TB of it This indicates 3995WX supports not only 8ch but also RDIMM/LRDIMM that's not supported on current Threadripper. It should need a new Socket rather than current TRX40 (TRX80 was rumored a year ago). Threadripper gets closer to EPYC. I expect PCIe lanes limitation still remains for make differenc…

That would be insane. I hope motherboard manufacturers take notice and release an IPMI board for it. We currently run a cluster that has some 3970X's in it and it's super awkward that we couldn't get remote management for them. Asrock has announced availability of one starting august, but it's a little too late for us.

Re: Ryzen Threadripper Pro 3995WX Spotted

#48

Post by adamtulinius »

CSDude wrote: post #41
vosper wrote: post #40

Earlier quoted context omitted.

I’m not sure about “require”, but we use multi-hundred-gigabytes or RAM instances for our Elasticsearch coordinator nodes, which seems to work well.

Elasticsearch recommends <32GB because of 32-bit per their doc, is it different for coordinator nodes?

That recommendation is only for the JVM heap to my knowledge.

Re: Ryzen Threadripper Pro 3995WX Spotted

#49

Post by mschuetz »

Uehreka wrote: post #12

I feel like 64 cores is getting rather close to a tipping point: What workloads are so massively parallel that they can use 64 cores of x86 but can’t use the thousands of CUDA cores on a Quadro card? I’m sure right now there are workloads that just need particular x86 instructions, but that feels like a temporary problem. Am I wrong about that being a temporary problem (that would feel frustrating)? Are these cores j…

Two reasons why I can't do my stuff with CUDA even though I've got thousands of work packages that are independent of each other:

* I'm splitting data sets into batches of around 50-300MB, but each individual batch must be processed sequentially. Each batch loops through the data (multiple times), and every single iteration of a loop depends on state from previous iterations.

* Altough each batch essentially does the same thing, it does not operate in lockstep. Lot's of ifs, different execution paths, differently sized sub-problems, etc.

I'd benefit from having 200 CPU cores and enough RAM to fit the batches, but I can't let individual GPU threads run through batches of 50MB, each.

Re: Ryzen Threadripper Pro 3995WX Spotted

#50

Post by dpwm »

qayxc wrote: post #44
jacquesm wrote: post #42

Earlier quoted context omitted.

The people that use CUDA don't care about the underlying architecture, they need results. Which ever manufacturer will deliver those results will get the $. Raytracing is not an ideal match workload wise for a GPU, too much divergence, but there are plenty of workloads where GPUs will have an edge for along time to come.

It's more than that - NVIDIA invested massively in their software ecosystem and mindshare (e.g. by a huge sponsoring program that saw universities flooded with their hardware). AMD cards offer comparable compute performance and are sometimes even better. What AMD lacks, though, are ready-to-use software packages like CUDA, cuDNN, cuBLAS, and the like. NVIDIA basically bought their way into the research community and…

> a huge sponsoring program that saw universities flooded with their hardware

This is really important, and I don't think people quite realise the scale. I was doing a Physics MSc in 2012 and it seemed that almost everybody in computational physics was doing some variant of reimplementing Y with CUDA.

That's not to say it was the wrong tool for many of the jobs – though for some it apparently was, judging by the levels of stress – but I did find it a bit weird how this was an approach to research that pretty much completely ignored vendor lock-in.

Post reply on HN