Blog post: "Unified GPU/SMP benchmarking scheme ..."

[Jesse_V]

I just saw this on my RSS reader and thought that this blog post was a pretty significant announcement.
Dr. Pande has said that they'd like more GPU clients folding. This will definitely help:

Unified GPU/SMP benchmarking scheme: equal points for equal work

The current benchmarking calculations for SMP and GPU projects are performed on different machines since originally the SMP cores could not perform the calculations that the GPUs cores could and vice versa (GPUs were only for implicit solvent calculations and SMP only for explicit solvent calculations). With recent advances in both cores and completion of our testing of these capabilities to ensure agreement, we are now confident we can do the same work on both cores. Thus, we feel that it is time to unify GPU and SMP benchmarking, both for simplicity and fairness.

In order to complete the move towards this plan of "equal points for equal work," new GPU projects will be benchmarked using the existing SMP benchmarking scheme. Based on our internal tests, the end effect of this new, unified benchmarking scheme would boost the points for the GPU projects, both in terms of base points but also by bringing Quick Return Bonuses to GPU clients. In order to test the new scheme, we have started a GPU3 project (Project ID: 8057) and released it for beta testing. Once the benchmarking scheme has been tested, all the current GPU projects will be re-benchmarked to reflect the changes in the benchmarking scheme.

[jimerickson]

so this is a big step towards qrb for gpu?

[Nathan_P]

I'm assuming that this will not just be restricted to nvidia GPU's?

[Jesse_V]

so this is a big step towards qrb for gpu?

It would appear so.

I'm assuming that this will not just be restricted to nvidia GPU's?

ATI wasn't specifically mentioned, so I think that can be assumed AFAIK.

[SwingOp3]

Will these be available using V7 and the "advanced" tag?

[muziqaz]

As far as I understand it will be for closed beta 1st

[Jesse_V]

As far as I understand it will be for closed beta 1st

I think you mean Beta Team testers, like everyone in viewforum.php?f=66

[muziqaz]

As far as I understand it will be for closed beta 1st

I think you mean Beta Team testers, like everyone in viewforum.php?f=66

Yes. That is Closed Beta. Open beta is advmethods guys

[k1wi]

It looks like the unification between gpu/cpu is being driven by the improvements enabled by gromacs 4.6 so gromacs is certainly making strides.

I'm pretty confident the AMD issues are more due to the constantly moving goal posts and openCL (and its adoption by AMD) being much much younger than CUDA.

[Evil Penguin]

I'm assuming that this will not just be restricted to nvidia GPU's?

Probably, at least until AMD gets their stuff together with their OpenCL compiler.

[mdk777]

Probably, at least until AMD gets their stuff together with their OpenCL compiler.

Link ?

Tools exist(compilers,SDK, debugging suite) , doesn't mean they are being used.....

[Evil Penguin]

Sorry, meant probably will be restricted until... AMD gets their stuff together with their OpenCL compiler.

[k1wi]

I appreciate that everyone has seemed to have focused on what this will mean in terms of PPD, QRB and whether AMD or NVidia will 'get it', but I really think the second half of the first sentence and the second sentence mean a hell of a lot more for the science of folding@home:

originally the SMP cores could not perform the calculations that the GPUs cores could and vice versa (GPUs were only for implicit solvent calculations and SMP only for explicit solvent calculations). With recent advances in both cores and completion of our testing of these capabilities to ensure agreement, we are now confident we can do the same work on both cores.

[kiore]

I appreciate that everyone has seemed to have focused on what this will mean in terms of PPD, QRB and whether AMD or NVidia will 'get it', but I really think the second half of the first sentence and the second sentence mean a hell of a lot more for the science of folding@home:

originally the SMP cores could not perform the calculations that the GPUs cores could and vice versa (GPUs were only for implicit solvent calculations and SMP only for explicit solvent calculations). With recent advances in both cores and completion of our testing of these capabilities to ensure agreement, we are now confident we can do the same work on both cores.

I am also hopefulthat this will mean a huge leap in productivity and participation.. Adding a GPU is one of the easiest upgrades to existing systems and although the electricity usage has been high going the GPU route rather than multi cpus, I see the potential if this works out for encouraging people to upgrade, or throwing a GPU in that 'old boxen' and firing it up again. This could be a boost to hardware spending like Big Advanced was.. At the very least encouraging those who already have the hardware but are not folding on it.

[KWSN_PToT]

...
I am also hopeful that this will mean a huge leap in productivity and participation. Adding a GPU is one of the easiest upgrades to existing systems and although the electricity usage has been high going the GPU route rather than multi cpus, I see the potential if this works out for encouraging people to upgrade, or throwing a GPU in that 'old boxen' and firing it up again. This could be a boost to hardware spending like Big Advanced was.. At the very least encouraging those who already have the hardware but are not folding on it.

I think this will have a chilling effect on those with older equipment.
There are many dedicated folders (veterans of many years) who are not able to afford to upgrade their hardware every few months/years.
Now, there is an added dis-incentive: Early Return Bonus (for GPUs).

Some members of my team have expressed dismay at this move, and are seriously considering dropping F@H for BOINC/WCG.

As an alternative to Dr. Pande's suggestion of unification (which actually gives unfair advantage to newer/faster hdwe), why not give bonus points for "consistent" contribution (over some period of months or years)???

I happen to be the #1 point-holder on my team (at the time of this post) and I have made a significant investment (recently) in newer hardware.
And still, I think this move is manifestly unfair to others who have provided many years of contribution (but their wallets and/or life-priorities will not allow them to upgrade).

Sorry to be such a Negative-Ned, but I'm scoring the objectives as...
Simplification: Marginal PASS.
Fairness: FAIL.