Re: [gmx-users] Performance of GTX 980 and 970
On 29.09.2014 14:31, Pappu Kumar wrote: I am wondering if anyone tested the performance of new GTX 980 and 970 cards and compared to 780/780Ti/Titan using the input systems given here As long as I don't have a card, I can only guess. But: until the appearance of reasonable benchmarks, you can look at the F@H-results: single: http://images.anandtech.com/graphs/graph8526/67747.png double: http://images.anandtech.com/graphs/graph8526/67748.png M. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
On Mon, Sep 29, 2014 at 2:46 PM, Mirco Wahab wrote: > On 29.09.2014 14:31, Pappu Kumar wrote: >> >> I am wondering if anyone tested the performance of new GTX 980 and 970 >> cards and compared to 780/780Ti/Titan using the input systems given here > > > As long as I don't have a card, I can only guess. > > But: until the appearance of reasonable benchmarks, you > can look at the F@H-results: > > single: > http://images.anandtech.com/graphs/graph8526/67747.png Note that that's the GB single precision benchmark's plot which shows surprising large performance gap between 780 Ti and 980 - perhaps there's something in the OpenMM GB kernels that favors Maxwell. This is the explicit solvent performance plot: http://images.anandtech.com/graphs/graph8526/67746.png showing ~25% improvement over a 780 Ti (still pretty impressive). This is close to what I'd have guessed based on extrapolating the 750 Ti performance. Also note that this is OpenCL performance which is inherently worse than CUDA and the OpeCL-CUDA difference may not be the same across all architectures! Cheers, -- Sz. > double: > http://images.anandtech.com/graphs/graph8526/67748.png > > > M. > > > > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a > mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
Hi, On 29 Sep 2014, at 16:27, Szilárd Páll wrote: > On Mon, Sep 29, 2014 at 2:46 PM, Mirco Wahab > wrote: >> On 29.09.2014 14:31, Pappu Kumar wrote: >>> >>> I am wondering if anyone tested the performance of new GTX 980 and 970 >>> cards and compared to 780/780Ti/Titan using the input systems given here >> >> >> As long as I don't have a card, I can only guess. >> >> But: until the appearance of reasonable benchmarks, you >> can look at the F@H-results: >> >> single: >> http://images.anandtech.com/graphs/graph8526/67747.png > > Note that that's the GB single precision benchmark's plot which shows > surprising large performance gap between 780 Ti and 980 - perhaps > there's something in the OpenMM GB kernels that favors Maxwell. > > This is the explicit solvent performance plot: > http://images.anandtech.com/graphs/graph8526/67746.png > showing ~25% improvement over a 780 Ti (still pretty impressive). This > is close to what I'd have guessed based on extrapolating the 750 Ti > performance. Wouldn’t one expect the 980 to be slightly slower than the 780 Ti due to its smaller single precision processing power? Carsten > > Also note that this is OpenCL performance which is inherently worse > than CUDA and the OpeCL-CUDA difference may not be the same across all > architectures! > > Cheers, > -- > Sz. > >> double: >> http://images.anandtech.com/graphs/graph8526/67748.png >> >> >> M. >> >> >> >> -- >> Gromacs Users mailing list >> >> * Please search the archive at >> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! >> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists >> >> * For (un)subscribe requests visit >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a >> mail to gmx-users-requ...@gromacs.org. > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a > mail to gmx-users-requ...@gromacs.org. -- Dr. Carsten Kutzner Max Planck Institute for Biophysical Chemistry Theoretical and Computational Biophysics Am Fassberg 11, 37077 Goettingen, Germany Tel. +49-551-2012313, Fax: +49-551-2012302 http://www.mpibpc.mpg.de/grubmueller/kutzner http://www.mpibpc.mpg.de/grubmueller/sppexa -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
On Mon, Sep 29, 2014 at 7:32 PM, Carsten Kutzner wrote: > Hi, > > On 29 Sep 2014, at 16:27, Szilárd Páll wrote: > >> On Mon, Sep 29, 2014 at 2:46 PM, Mirco Wahab >> wrote: >>> On 29.09.2014 14:31, Pappu Kumar wrote: I am wondering if anyone tested the performance of new GTX 980 and 970 cards and compared to 780/780Ti/Titan using the input systems given here >>> >>> >>> As long as I don't have a card, I can only guess. >>> >>> But: until the appearance of reasonable benchmarks, you >>> can look at the F@H-results: >>> >>> single: >>> http://images.anandtech.com/graphs/graph8526/67747.png >> >> Note that that's the GB single precision benchmark's plot which shows >> surprising large performance gap between 780 Ti and 980 - perhaps >> there's something in the OpenMM GB kernels that favors Maxwell. >> >> This is the explicit solvent performance plot: >> http://images.anandtech.com/graphs/graph8526/67746.png >> showing ~25% improvement over a 780 Ti (still pretty impressive). This >> is close to what I'd have guessed based on extrapolating the 750 Ti >> performance. > Wouldn’t one expect the 980 to be slightly slower than the 780 Ti > due to its smaller single precision processing power? You are referring to the decreased (theoretical) instruction throughput per multiprocessor of by 1.5x, right? Considering only this aspect, you are right, one would expect lower performance. However, the Maxwell architecture seems to be quite a bit more efficient than Kepler, in particular I think it has become much better at instruction scheduling (as advertised). That's not such a surprise as there was quite some room for improvement, I think. The theoretical IPC of 7 on Kepler was like a dream that never materialized, even codes with the right mix SP instructions would struggle to get above an IPC of 5-6. I have not profiled code on Maxwell extensively (yet), but what I know from having compared our Ewald non-bonded kernel's peak throughput on the 750 Ti vs 780 Ti is that the clock to clock* per multiprocessor throughput is only ~15% higher on the 780 Ti which is much less than the decrease in instruction throughput. Additionally, the GTX 980 has and extra multiprocessor (16 iso 15) and higher base frequency than the 780 Ti to compensate. *Considering base clocks is not entirely correct as the stable boost clock delta might differ. -- Szilárd > Carsten > >> >> Also note that this is OpenCL performance which is inherently worse >> than CUDA and the OpeCL-CUDA difference may not be the same across all >> architectures! >> >> Cheers, >> -- >> Sz. >> >>> double: >>> http://images.anandtech.com/graphs/graph8526/67748.png >>> >>> >>> M. >>> >>> >>> >>> -- >>> Gromacs Users mailing list >>> >>> * Please search the archive at >>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! >>> >>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists >>> >>> * For (un)subscribe requests visit >>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a >>> mail to gmx-users-requ...@gromacs.org. >> -- >> Gromacs Users mailing list >> >> * Please search the archive at >> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! >> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists >> >> * For (un)subscribe requests visit >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a >> mail to gmx-users-requ...@gromacs.org. > > > -- > Dr. Carsten Kutzner > Max Planck Institute for Biophysical Chemistry > Theoretical and Computational Biophysics > Am Fassberg 11, 37077 Goettingen, Germany > Tel. +49-551-2012313, Fax: +49-551-2012302 > http://www.mpibpc.mpg.de/grubmueller/kutzner > http://www.mpibpc.mpg.de/grubmueller/sppexa > > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a > mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
The 6-core Intel CPUs have only 28 PCI-E lanes rather than 40 like the 5830K/5860X which means that with a second GPU you'll get x16/x8 and with three GPUs x8/x8/x8. Also note that for the current GROMACS implementation, pairing a 5820K with two 980s will likely give a rather imbalanced hardware setup - with three 970s even more so (at least for common types of run setups). Depending on the exact use case, you may be able to make good use of 2-3 GPUs even with just a 5820K (e.g. in multi runs, one per GPU) or setups with long cut-off (or without PME), but otherwise you may not see much benefit from a second GPU, let alone a third. -- Szilárd On Mon, Sep 29, 2014 at 5:01 PM, Pappu Kumar wrote: > Thank you for your info. I am planning to buy a computer with the following > configuration: > > Intel 5820K > Corsair H100i Hydro Cooling Performance > MSI X99 SLI Plus > Fractal Design R4 > Seasonic X 1050W > > I am wondering if it would be a good idea to go for 3x GTX 970 instead of 2x > GTX 980 since the cost is the same. Thank you. > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a > mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
I am planning to buy 2x GTX 970 for 5820K overclocked to 4.5 GHz. I have budget limitations and not able to afford workstations with 2x CPUs. Let me know if you are aware of any cheaper alternative. Also let me know if in future Gromacs could become more GPU intensive allowing more GPUs with one CPU. Thank you. On Tuesday, 30 September 2014 5:18 PM, Szilárd Páll wrote: The 6-core Intel CPUs have only 28 PCI-E lanes rather than 40 like the 5830K/5860X which means that with a second GPU you'll get x16/x8 and with three GPUs x8/x8/x8. Also note that for the current GROMACS implementation, pairing a 5820K with two 980s will likely give a rather imbalanced hardware setup - with three 970s even more so (at least for common types of run setups). Depending on the exact use case, you may be able to make good use of 2-3 GPUs even with just a 5820K (e.g. in multi runs, one per GPU) or setups with long cut-off (or without PME), but otherwise you may not see much benefit from a second GPU, let alone a third. -- Szilárd On Mon, Sep 29, 2014 at 5:01 PM, Pappu Kumar wrote: > Thank you for your info. I am planning to buy a computer with the following > configuration: > > Intel 5820K > Corsair H100i Hydro Cooling Performance > MSI X99 SLI Plus > Fractal Design R4 > Seasonic X 1050W > > I am wondering if it would be a good idea to go for 3x GTX 970 instead of 2x > GTX 980 since the cost is the same. Thank you. > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a > mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Performance of GTX 980 and 970
On Wed, Oct 1, 2014 at 5:40 PM, Pappu Kumar wrote: > I am planning to buy 2x GTX 970 for 5820K overclocked to 4.5 GHz. I have > budget limitations and not able to afford workstations with 2x CPUs. Let me > know if you are aware of any cheaper alternative. Sounds like a good investment! The only case where 2x970 could perform worse than a single 980 is if your input system is quite small and even that should change in future versions. > Also let me know if in future Gromacs could become more GPU intensive > allowing more GPUs with one CPU. Thank you. Certainly! The bonded interactions will certainly be offloaded in the near future! Cheers, -- Szilárd > > On Tuesday, 30 September 2014 5:18 PM, Szilárd Páll > wrote: > > > The 6-core Intel CPUs have only 28 PCI-E lanes rather than 40 like the > 5830K/5860X which means that with a second GPU you'll get x16/x8 and > with three GPUs x8/x8/x8. > > Also note that for the current GROMACS implementation, pairing a 5820K > with two 980s will likely give a rather imbalanced hardware setup - > with three 970s even more so (at least for common types of run > setups). Depending on the exact use case, you may be able to make good > use of 2-3 GPUs even with just a 5820K (e.g. in multi runs, one per > GPU) or setups with long cut-off (or without PME), but otherwise you > may not see much benefit from a second GPU, let alone a third. > > -- > Szilárd > > > On Mon, Sep 29, 2014 at 5:01 PM, Pappu Kumar wrote: >> Thank you for your info. I am planning to buy a computer with the >> following configuration: >> >> Intel 5820K >> Corsair H100i Hydro Cooling Performance >> MSI X99 SLI Plus >> Fractal Design R4 >> Seasonic X 1050W >> >> I am wondering if it would be a good idea to go for 3x GTX 970 instead of >> 2x GTX 980 since the cost is the same. Thank you. > >> -- >> Gromacs Users mailing list >> >> * Please search the archive at >> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! >> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists >> >> * For (un)subscribe requests visit >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send >> a mail to gmx-users-requ...@gromacs.org. > -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.