GROMACS in Spec 2006

Moderators: Site Moderators, FAHC Science Team

Post Reply
7im
Posts: 10179
Joined: Thu Nov 29, 2007 4:30 pm
Hardware configuration: Intel i7-4770K @ 4.5 GHz, 16 GB DDR3-2133 Corsair Vengence (black/red), EVGA GTX 760 @ 1200 MHz, on an Asus Maximus VI Hero MB (black/red), in a blacked out Antec P280 Tower, with a Xigmatek Night Hawk (black) HSF, Seasonic 760w Platinum (black case, sleeves, wires), 4 SilenX 120mm Case fans with silicon fan gaskets and silicon mounts (all black), a 512GB Samsung SSD (black), and a 2TB Black Western Digital HD (silver/black).
Location: Arizona
Contact:

GROMACS in Spec 2006

Post by 7im »

Wednesday, 12 September 2007
The GROMACS code is now part of the SPEC 2006 benchmark system for CPUs . That means hardware vendors that want to show off their hardware run GROMACS as a part of their benchmarks and some of them tune hardware and/or compilers to get better results. Look for test 435.gromacs to compare your favorites hardware. Note though, that for the benchmark we were not allowed to use assembly code, so the results you see there are in C-code. As a result the numbers are not representative of true GROMACS performance.
Gromacs News
SPEC Site

So it sounds like we will start getting some relative performance numbers from the AMD and Intel camps, as it relates to Gromacs, but now the question is how closely this relates to actual client performance.

For instance, I have seen numbers like this posted, but don't know how well they translate...

435.gromacs, molecular dynamics, lower is better
Dual Opteron 2350 2 GHz (8 cores) Base: 752 s Peak: 607 s
Dual Xeon 5345 2.33 GHz (8 cores) Base: 606 s Peak: 590 s

The base score makes the AMD look ~40% slower, but the peak numbers are much closer. Any help translating this info to common terms?
Last edited by 7im on Wed Dec 19, 2007 4:16 pm, edited 1 time in total.
Reason: add linky
Post Reply