PC GAMESS v. 7.0.2 Kentsfield benchmarks


Number of cores used

1

2

3

4

Test 1,

CPU time and relative speedup

2936.2

100%

1488.0

197%

1033.9

284%

804.9

365%

Test 2,

Wall clock time and relative speedup

164.9

100%

111.2

148%

93.5

176%

85.5

193%

Test 3,

CPU time and relative speedup

5571.9

100%

2713.7

205%

1834.9

304%

1383.4

403%

Test 4,

Wall clock time and relative speedup

713.0

100%

378.4

188%

292.2

244%

222.1

321%

Test 5,

CPU time and relative speedup

5929.2

100%

3105.2

191%

2258.7

263%

1867.9

317%

Test 6,

CPU time and relative speedup

16817.6

100%

8461.1

199%

5751.8

292%

4399.9

382%

Standard MP4(SDTQ) benchmark,

Wall clock time and relative speedup

5628.7

100%

2899.0

194%

2065.4

273%

1607.0

350%

 


OS and hardware description


Intel Quad-core Core 2 Quadro QX-6700 (Kentsfield) 2.67 GHz, 1067 MHz FSB, Asus P5B (Vanilla) board, 4x 1GB DDR-2 800 5-5-5 DIMMs, 2x Seagate Barracuda-10 320 GB SATA HDDs configured as software RAID-0, Opensuse 10.2 64-bit.




Tests description


Test 1, single-point direct DFT (B3LYP) energy plus gradient for medium-size system (623 basis functions). View image

Test 2, single-point semiempirical (PM3) energy plus gradient for large system (540 atoms, 2160 basis functions). View image

Test 3, single-point direct MP2 energy for medium-size system (623 basis functions, the same system as one used for Test 1). View image

Test 4, single-point two-state MCQDPT2 energy with ISA energy denominators shift for small model system. View image

Test 5, single-point direct CASSCF(12,12) for medium-size system (retinal molecule, cc-pVDZ, 565 Cartesian basis functions) using ALDET code. View image

Test 6, single-point direct CIS energy plus gradient of first excited state of medium-size system (porphyrin molecule, cc-pVTZ (aug-cc on Nitrogens), 1130 Cartesian basis functions, D2h group). View image

More data on standard MP4(SDTQ) benchmark


Test comments


Tests 2, 4, as well as standard MP4 benchmark were run in multithreaded mode, other tests were run in standard parallel mode using dynamic load balancing over p2p interface. Call64 switch was turned on for all tests for faster processing. Note that test 2 does not scale well mainly due to limitations of the PC GAMESS' semiempirical code, while test 4 would scale much better for larger job. CPU or Wall clock times are given on master node in seconds.


Copyright © 2006 by Alex A. Granovsky

We are grateful to Dr. Peter Burger who kindly provided us by the access to hardware

Press to visit PC GAMESS' Woodcrest vs. Opteron performance comparison page

Press to visit PC GAMESS Pentium 4 family benchmarks page to compare the results of these benchmarks with those obtained on various Netburst (Pentium 4 and Pentium D) processors.

Press to return back to the PC GAMESS vs. WinGamess performance comparison page to compare the results of these benchmarks with those obtained on other processors