Message boards : Cafe Rosetta : Record performances with EPYC
Author | Message |
---|---|
[VENETO] boboviz Send message Joined: 1 Dec 05 Posts: 1994 Credit: 9,596,782 RAC: 8,757 |
Supermicro Sets World Record Performance with AMD EPYC Processors There are the new cpus Epyc 7fx2 These cpus on Rosetta (maybe with SSEx/Avx instrunctions!!) will be great. :-P |
spRocket Send message Joined: 23 Mar 20 Posts: 22 Credit: 3,008,018 RAC: 0 |
*drool* I wish I could afford even one of those. Even a Ryzen 9 is a bit too rich for my finances, though. |
Grant (SSSF) Send message Joined: 28 Mar 20 Posts: 1679 Credit: 17,802,537 RAC: 22,591 |
*drool*Yeah. I used a web site to set up a new system (just in case i won the lotto) based on a Ryzen 9 3900X. $3,000 and a bit (and that's without a video card, i'd use one from an existing system). Grant Darwin NT |
William Albert Send message Joined: 22 Mar 20 Posts: 23 Credit: 1,069,070 RAC: 147 |
The AMD EPYC 7Fx2 series is intended for applications that benefit more from frequency and cache, over raw thread count. However, Rosetta@home is perfectly able to use as many threads as a system has available (provided that the machine has sufficient RAM). While these frequency-optimized SKUs are impressive for their intended workload, they wouldn't be optimal for Rosetta compared to the existing 2nd generation Epyc lineup. |
Chilean Send message Joined: 16 Oct 05 Posts: 711 Credit: 26,694,507 RAC: 0 |
The AMD EPYC 7Fx2 series is intended for applications that benefit more from frequency and cache, over raw thread count. However, Rosetta@home is perfectly able to use as many threads as a system has available (provided that the machine has sufficient RAM). Isn't Rosetta L3-Cache hungry? Is it correct that as time goes on, L3 cache will eventually turn into "RAM"? I remember having a computer with 64MB of RAM... and now we have a CPU with 256MB of L3 CACHE. |
Mod.Sense Volunteer moderator Send message Joined: 22 Aug 06 Posts: 4018 Credit: 0 RAC: 0 |
Any large application, or application that uses large data structures will be "cache hungry", and will perform better with a larger cache available. Rosetta Moderator: Mod.Sense |
Message boards :
Cafe Rosetta :
Record performances with EPYC
©2024 University of Washington
https://www.bakerlab.org