++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
mem wall CPUs compared run
machine model MHz MB comp hr:min /step to aresAlpha date
--------- -------------- ---- --- ----- ----- ------ ------------ -----
mars.math Intel Xeon 3600 24GB gfor 0:06 0.0108 22 x faster jun15
ares.math Intel Xeon 3600 16GB gfor 0:06 0.0109 22 x faster aug15
householder Intel Xeon 3000 192GB gfor 0:07 0.0135 18 x faster may14
frost SGI Altix8200 2400 2GB ifort 0:07 0.0134 18 x faster mar10
zeus AMD quad-2376 2300 2GB ifort 0:09 0.0178 13 x faster feb10
midtown AMD quad-2376 2300 2GB ifort 0:09 0.0184 13 x faster feb10
zeus AMD opteron252 2600 2GB pgf95 0:12 0.0229 10 x faster apr06
newton Intel Xeon 3192 4GB ifort 0:14 0.0264 9 x faster may06
tiger Cray opteron248 2200 pgf95 0:14 0.0264 9 x faster dec07
oic Intel Xeon 3391 4GB ifort 0:14 0.0271 8.7x faster may06
fubini Intel Xeon 3056 4GB ifort 0:17 0.0315 7.7x faster oct03
abcd Intel Xeon 3189 4GB ifort 0:17 0.0331 7.1x faster oct04
hawk opteron 242 1600 2GB pf90 0:20 0.0374 6.4x faster jan05
tiger Cray opteron248 2200 g77 0:20 0.0377 6 x faster dec07
zeus AMD opteron252 2600 2GB g77 0:21 0.0397 6.0x faster apr06
agnesi Intel Xeon 2187 4GB ifort 0:23 0.044 5.4x faster nov02
ares(new) Intel Core2 1860 2GB gfortr 0:25 0.0469 5.4x faster jul07
hawk AMD opteron242 1600 2GB g77 0:26 0.048 5.5x faster jan05
colt Alpha SC ev67 667 2GB f90 0:28 0.0526 4.5x faster apr01
frodo AMD opteron240 1400 2GB g77 0:29 0.053 4.4x faster oct04
abcd Intel Xeon 3189 4GB g77 0:32 0.060 4.0x faster oct04
cheetah IBM Pwr4(p690) 1300 1GB xlf 0:32 0.0614 3.8x faster jul02
oic Intel Xeon 3391 4GB g77 0:35 0.0671 3.6x faster may06
newton Intel Xeon 3192 4GB g77 0:38 0.0721 3.3x faster may06
animal Alpha ev6(21264) ? ? f77 0:39 0.0726 3.3x faster dec99
knox3 Sun UltraSparc 900 1GB f77 1:00 0.1136 2.0x faster jan05
eagle IBM SP Wnthawk 375 ? xlf 1:02 0.1174 2.0x faster jan01
mulato Alpha PC/500au 500 256 f77 1:21 0.154 1.6x faster oct97
barnard Sun ultra80 450 1GB f77 1:30 0.1705 1.6x faster mar01
apollonius Alpha/Linux 533 216 g77 1:30 0.1458 1.6x faster feb00
vxa Dell LatitudeC600 752 261 g77 1:33 0.1752 1.3x faster aug01
power3 IBM Pwr3 dual 200? ? xlf 1:45 0.197 1.2x faster nov99
torc Intel P4 dual 550 256 pgf90 1:48 0.197 1.2x faster may00
macho SGI64 R10000 ? 512 f77 1:52 0.20 1.2x faster oct96
zeus Alpha 500 333 64 f77 1:51 0.2087 1.1x faster dec99
cauchy Intel P4 dual 600 128 g77 2:00 0.2265 1.1x faster may00
ares Alpha 500 ev6 333 64 f77 2:06 0.24 1.0x oct96
nala Sun ultra2 140 128 f77 3:29 0.4 1.7x slower oct97
f2n7 IBM SP2 PWR2 120 256 xlf 3:44 0.42 1.7x slower may97
baloo0 IBM RS6000 590 66 ? xlf 6:50 0.78 3.3x slower jan95
nautique Alpha 2100? 233? ? f77 8:40 0.94 3.9x slower jan95
mathsun27 Sun sparc 20x51 50 32 f77 12:25 1.40 5.8x slower jan95
austin IBM RS6000 550 66 ? xlf 13:47 1.54 6.4x slower dec94
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
If you would be willing to run the benchmark on another machine
teal.epm.ornl.gov
IBM RS6000/590 66MHz 128MB 256+32K cache 4dec94
25063.0u 0.0s 17:40:31 39%
==> 0.79 CPUs/step in 17:40 hours
good CPU speed but terrible wall clock time!
austin.cs.utk.edu 4dec94
IBM RS6000/550 66MHz ?MB
uname -a: AIX austin 1 4 000005721C00
xlf -O3 -qstrict -qnolm -qtune=pwr2 -qhot
48553.0u 0.0s 13:46:59 97%
==> 1.54 CPUs/step in 13:47 hours
python.cs.utk.edu 4dec94
HP 750
f77 -O3 ?
59268.5u 12.1s 29:04:24 56%
==> 1.88 CPUs/step in 29 hours !!! terrible !!!
nautique.epm.ornl.gov 4dec94
Digital Alpha 2100? ev4(21064) ?233MHz?
29670.07u 486.32s 8:42:10 96%
==> 0.94 CPUs/step in 8:40 hours !!!
baloo0.epm.ornl.gov 14jan95
IBM RS6000/590/66MHz
AIX ???: xlf ???
24600.560u 0.040s 6:50:28.11 -74.-5%
==> 0.78 CPUs/step in 6:50 hours
mathsun27.math.utk.edu 16jan95
Sun sparc20/51 50MHz, 32MB, 1M cache
f77 SC3.0.1: f77 -fast -O4 -xcg92
44442.780u 3.300s 12:25:02.94 3.3%
==> 1.40 CPUs/step in 12:25 hours
manzana.epm.ornl.gov 16jan95
SGI 5/ irix 5.2 IP22
61321.6u 1480.0s 17:45:22 98%
==> 1.94 CPUs/step in 17:45 hours !!! terrible !!!
mathsun33.math.utk.edu 3may95
SGI Indigo 2 XZ R4400/200MHz, 64Mb
28549.552u 645.744s 8:12:01.28 -46.-5%
==> 0.90 CPUs/step in 8:12 hours
nala.cs.utk.edu 18nov95
Sun Ultrasparc 1, 140MHz, 128Mb ; Solaris ???
f77 -fast -O4
18834.4u 0.3s 5:14:12.5
a600.aitcorp.com 13dec95
Digital Alpha 5/300 300MHz, ??Mb
f77 -fast -O5 -tune ev5
9790.43u 0.18s 2:43:19 99%
ares.math.utk.edu 18oct96
Digital Alpha 500/333MHz, 64Mb while working!
uname -a: OSF1 ares.math.utk.edu V4.0 386 alpha
DEC Fortran 90 V1.3: f90 -fast -O5 -tune ev5
7512.929u 6.623s 2:05:40.90 99.7%
==> 0.238 CPUs/step in 2:06 hours!
macho.epm.ornl.gov 18oct96
SGI64 R10000 4-proc SMP, 512Mb?
uname -a: IRIX64 macho 6.2 03131016 IP25
f77 -Ofast -O4
6308.954u 345.004s 1:51:50.91 99.1%
==> 0.20 CPUs/step in 1:52 hours!
f3n1.cas.utk.edu 4may97
IBM SP2 highnode: 8-112MHz RS6K POWERPC 604 1Gb
uname -a: AIX f3n1 1 4 000852A8A400
f77 -O3 -qarch=pwr2/pwrx -qtune=pwr2/pwrx
24555.430u 0.170s 6:49:18.21 -74.-8%
==> 0.77 CPUs/step in 6:49 hours
f2n7.cas.utk.edu 4may97
IBM SP2 thinnode: 120MHz POWER2 SC, 256Mb
uname -a: AIX f2n7 1 4 000034188100
f77 -O3 -qarch=pwr2/pwrx -qtune=pwr2/pwrx
13312.210u 0.060s 3:44:11.44 98.9%
==> 0.42 CPUs/step in 3:44 hours
nala.cs.utk.edu 11oct97
Sun Ultra-2
uname -a: SunOS nala 5.5.1 Generic sun4u sparc SUNW,Ultra-2
12566.66u 0.05s 3:29:35.22 99.9%
==> 0.4 CPUs/step in 3:29 hours
blueberry.cs.utk.edu 11oct97
SGI ???
uname -a: IRIX blueberry 5.3 11091812 IP22 mips
f77 -O4,3 did not compile, used f77 -O2
70253.720u 990.846s 20:35:30.15 -19.-7%
worst ever !!! must be an old machine!
picasso.cs.utk.edu 11oct97
SGI64 R10000 ?
uname -a: IRIX64 picasso 6.2 06101031 IP28
f77 -Ofast
6351.055u 396.142s 1:54:05.60 98.5%
==> 0.2 CPUs/step in 1:54 hours. Pretty good, like macho!
mulato.epm.ornl.gov 11oct97
Digital Alpha PC/500MHz, 256Mb ; Unix V4.0
uname -a: OSF1 mulato.epm.ornl.gov V4.0 564.32 alpha
no f77, compiled on zeus: f77 -fast -O5 -tune ev5
4869.143u 1.327s 1:21:16.13 99.8%
==> 0.154 CPUs/step in 1:21 hours !!! best so far !!!
power3.cs.utk.edu 20nov99
IBM Power3 dual SMP
uname -a: AIX power3 3 4 00005F6B4C00
xlf -O4 -qarch=auto -qnolm -qtune=pwr3
6262.840u 0.130s 1:45:08.34 99.2%
==> 0.197 CPUs/step in 1:45 hours
animal.cs.utk.edu 5dec99
Digital Alpha 21264 ev6, 256Mb?
uname -a: OSF1 animal V4.0 1091 alpha
DIGITAL Fortran 90 V5.2-705: f90 -fast -O5 -tune ev5
2303.177u 0.774s 38:34.00 99.5%
==> 0.0726 CPUs/step in 38.5 minutes !!! unbeatable !!!
f90 -fast -O5 -tune ev6
2319.066u 0.057s 38:40.98 99.9%
f90 -fast -O5 -arch host -tune host
2315.695u 0.753s 38:42.48 99.7% 0+10k 79+89io 4pf+0w
strange, -tune ev5 faster than -tune ev6
zeus.math.utk.edu 5dec99
Digital Alpha 21164 ev5, 64Mb
uname -a: OSF1 zeus.math.utk.edu V4.0 878 alpha
DIGITAL Fortran 90 V5.2-705: f90 -fast -O5 -tune ev5
6621.589u 4.909s 1:51:29.23 99.0%
==> 0.2087 CPUs/step in 1:51 hours same as macho
apollonius.math.utk.edu 11feb00
Digital Alpha 533MHz ev5, 216MB
uname -a: Linux apollonius 2.0.35 1998 alpha unknown
Compaq Fortran Linux Alpha v1.0: fort -fast -O5 -tune ev5
4605.122u 1.917s 1:29:55.10 85.3%
==> 0.1458 CPUs/step in 1:30 hrs, not bad! beats power3 !!!
colt.ccs.ornl.gov on one CPU may00, oct00
Compaq AlphaServer SC, 4 SMP CPUs per node, 2GB RAM
CPU: ES40 processor: 21264a (ev67), 667 MHz,
64KB I-cache, 64KB D-cache, 8MB L2 cache
uname -a: OSF1 colt0 V5.0 910 alpha
f90 5.3: f90 -fast -O5 -tune ev6
1705.87u 0.06s 28:27 99% 0+10k 9+9io 1pf+0w 19oct00
==> 0.0537 CPUs/step in 28 MINUTES ! unreal !
On colt13 (with prun, no DFS/DCE) 25apr01
uname -a: OSF1 colt0 V5.1 732 alpha
f95 Compaq Fortran Compiler V5.4A-1472-46B2F
1669.84u 0.12s 27:51 99% 0+10k 158+0io 18pf+0w
On colt (with prun, from PFS) 28jul02
f90 5.3: f90 -fast -O5 -tune ev67
1666.31u 0.06s 27:47 99% 0+10k 42+0io 5pf+0w
==> 0.0525 CPUs/step in under 28 MINUTES !
cauchy.math.utk.edu 17may00
Gateway E-5200, dual PentiumIII, 600MHz, 128Mb
uname -a: Linux cauchy 2.2.12-20smp i686 unknown
g77 -O
7186.870u 5.280s 1:59:53.72 99.9%
==> 0.2265 CPUs/step in 2 hrs
torc0.cs.utk.edu 25may00
Intel ??? dual PentiumIII 550MHz 256MB 512cache
uname -a: Linux torc0 2.2.14 #1 SMP i686 unknown
GNU F77 version egcs-2.91.66 19990314/Linux
i386-redhat-linux compiled by GNU C version egcs-2.91.66
f77 -O
7364.190u 18.570s 2:03:31.38 99.6%
==> 0.2321 CPUs/step in 2:03 hrs
PGI pgf90 3.1-3: pgf90 -fast 29may00
6240.230u 171.340s 1:48:44.92 98.2%
==> 0.1966 CPUs/step in 1:48 hrs
eagle.ccs.ornl.gov (Pat Worley ran it) 11jan01
IBM SP 4-way Winterhawk II SMP nodes
375 MHz Power3-II processors with 8MB L2 cache
uname -a: AIX eagle163s 3 4 000101454C00
xlf -O3 -qstrict -qtune=pwr3 -qarch=pwr3 -qnolm -qhot -qipa -qfloat=hsflt
3725.4u 0.0s 1:02:05 99% 115+907k 0+0io 19pf+0w 11jan01
==> 0.1174 CPUs/step in 1:02 hrs
xlf_r -g -O4 -qnoipa on eagle164s from /tmp/gpfs200a/vxa/
4634.1u 24.8s 1:17:40 99% 99+1171k 0+0io 340pf+0w 28jul02
==> 0.146 CPUs/step in 1:18 hrs
xlf_r -g -O4 -qnoipa via on eagle164s LoadLeveler 29jul02
4538.10 0.26 1:15:39
==> 0.143 CPUs/step in 1:16 hrs
barnard.math.ua.edu (N.Hannoun ran it) 25mar01
Sun ultra-80 dual SMP 450MHz, 1GB, solaris 5.8
SunOS barnard 5.8 Generic_108528-06 sun4u sparc SUNW,Ultra-80
f95: Sun WorkShop 6 update 1 Fortran 95 6.1 2000/09/11
f95 -fast -O4
5411.0u 0.0s 1:30:13 99% 0+0k 0+0io 0pf+0w
==> 0.1705 CPUs/step in 1:30 hrs
vxa.math.utk.edu Dell Latitude C600 752MHz 261MB 14aug01
redhat7.1 linux2.4.2
g77 version 2.96 20000731 (RedHat Linux 7.1.2.96-81)
g77 -O3
5560.740u 1.690s 1:33:13.75 99.4% 0+0k 0+0io 146pf+0w
==> 0.1752 CPUs/step in 1:33 hrs
cheetah.ccs.ornl.gov 25jul02
IBM pSeries System (p690)
27 "Regatta" nodes, each with 32 processors on 16 chips
CPU: 1.3 GHz Power4 processor,
64 KB L1 cache, 32 KB D-cache, 1.5 MB L2 cache
estimated computational power 4.5 TeraFLOP/s
OS: AIX 5.1.0.0 uname -a: AIX cheetah0033 1 5 00207D8A4C00
Fortran level: 7.1.1.3
xlf_r -g -O4 -qnoipa
on cheetah0033 (login node) from /tmp/gpfs750a/vxa/: 25jul02
1949.020u 0.240s 32:30.27 99.9% 139+2215k
==> 0.0614 CPUs/step in 32 minutes
on cheetah1569 (compute node?) from /tmp/gpfs750a/vxa/: 28jul02
5761.2u 1.7s 1:32:38 103% 107+2198k
==> 0.182 CPUs/step in 1:33 minutes !!! why so slow ???
on cheetah0033 (login node) from /tmp/gpfs750a/vxa/: 28jul02
1956.28 0.02 32:36
==> 0.0617 CPUs/step in 32 minutes
agnesi.math.utk.edu:
dual Intel Pentium 4 XEON 2.2GHz 512KB cache, 4GB mem
uname -a: Linux 2.4.9-31enterprise Red Hat Linux release 7.2
g77 version 2.96 20000731 (Red Hat Linux 7.1 2.96-98)
g77 -O3
3025.680u 0.170s 50:25.66 15nov02
Intel(R) Fortran Compiler for 32-bit applications,
Version 6.0 Build 020312Z trial nov02
ifc -O3 -mp1 -tpp7
1437.090u 0.010s 23:57.10 dramatically better than g77 15nov02
==> 0.045 CPUs/step in 24 minutes ! beats the alpha !
ifc -O3 -tpp7
1385.700u 0.050s 23:05.67 slightly faster w/out -mp1
==> 0.044 CPUs/step in 23 minutes ! beats the alpha !
fubini.math.utk.edu
dual Intel Pentium 4 XEON 3.06GHz 512KB cache, 4GB mem
uname -a: Linux 2.4.20-20.9bigmem #1 SMP
gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)
g77 -O3
1840.870u 0.010s 30:41.38 13oct03
Intel(R) Fortran Compiler for 32-bit applications,
Version 7.1 Build 20030909Z
ifc -O3 -tpp7
997.740u 1.020s 16:38.82 99.9% unreal !!! 13oct03
==> 0.0314 CPUs/step in less than 17 minutes ! beats everything !
abcd.math.vanderbilt.edu
dual Intel Pentium 4 XEON 3.20GHz 512KB cache 4GB mem
uname -a: 2.4.9-e.3smp #1 SMP i686 unknown
gcc -v: gcc version 2.96 20000731 (Red Hat Linux 7.2 2.96-124.7.2)
g77 -O3 or -O5
1894.300u 0.000s 31:34.29 100.0% 27oct04
==> 0.0597 CPUs/step in 31.5inutes
ifort -v: Version 8.0
ifort -O3 -tpp7 -w95 -FI
1051.840u 0.000s 17:31.92 99.9% 27oct04
==> 0.0331 CPUs/step in 17.5 minutes !!! great !!!
frodo.sinrg.cs.utk.edu
dual AMD Opteron 240 1.4GHz 1024KB cache 2GB mem
uname -a: Linux head 2.4.19-NUMA #1 SMP x86_64
gcc -v: gcc version 3.2.2 (SuSE Linux)
g77 -O3
1695.330u 15.870s 28:33.94 99.8% on head node 27oct04
==> 0.0534 CPUs/step in 28.5 minutes
knox3.rgrid.utk.edu (node of knox OIT cluster)
Sun UltraSparc 900MHz 1MB
uname -a: SunOS knox1 5.9 Generic_112233-11 sun4u sparc SUNW,Sun-Fire-280R
f95 -V: Forte Developer 7 Fortran 95 7.0 2002/03/09
f95 -fast -O4
3605.0u 0.0s 1:00:20 99% 09jan05
==> 0.11362 CPUs/step in 1 hr
hawk.csm.ornl.gov (node of render hawk cluster)
dual AMD Opteron 242 1.6GHz 1024KB cache 2GB mem
g77 -v: gcc version 3.3.3 (SuSE Linux)
g77 -O3 -fno-automatic 1748.395u 0.923s 29:14.49
g77 -O3 -fPIC 1650.417u 0.166s 27:32.58 99.8%
g77 -O4 1542.776u 0.238s 25:45.56 99.8%
g77 -O3 1536.186u 0.051s 25:37.67 99.9% 23jan05
==> 0.048 CPUs/step in 26 min
g77 -v: gcc version 3.4.2
g77-3.4.2 -O3 1737.945u 0.713s 29:03.93 31jan05
g77-3.4.2 -O3 -fPIC 1623.384u 0.230s 27:05.85 1feb05
ifort Version 8.1
ifort -O3 1542.197u 1.056s 25:49.09 99.6% 23jan05
ifort -O4 1537.858u 0.460s 25:41.58 99.7%
==> 0.049 CPUs/step in 26 min
pathscale EKO compiler Suite(TM): Version 1.4
gcc version 3.3.1 (PathScale 1.4 driver)
pathf90 -O3 -mtune=opteron 1319.068u 0.212s 22:01.62
pathf90 -Ofast 1187.694u 0.856s 19:53.01
pathf90 -Ofast -fpic -mtune=opteron 1180.756u 0.075s 19:42.24
pathf90 -Ofast -mtune=opteron 1179.866u 0.020s 19:48.56 < best
==> 0.037 CPUs/step in 20 min 29jan05
zeus.math.utk.edu 9+headnode Opteron 252 linux cluster
dual AMD Opteron 252 2.6GHz 1024KB cache 2GB mem
uname -a: Linux 2.6.12-1.1381_FC3smp x86_64 GNU/Linux
g77 -v: gcc version 3.4.4 20050721 (Red Hat 3.4.4-2)
g77 -O3 1258.495u 0.023s 20:59.05 99.9% 14apr06
==> 0.037 CPUs/step in 21 min
pgf95 -V: pgf95 6.1-3 64-bit target on x86-64 Linux 14apr06
pgf95 -fast -O3 736.200u 0.010s 12:16.48
pgf95 -fast -O3 -Mcache_align 727.331u 0.007s 12:07.62
==> 0.0229 CPUs/step in 12 min !!! <-- best ever on fspx !!!
oic.ornl.gov 325 node Xeon linux cluster
dual Intel Xeon 3.4GHz 2048KB cache 4GB mem
uname -a: Linux b06l02 2.6.9-22.0.2.ELsmp #1 SMP x86_64 GNU/Linux
g77 -v: gcc version 3.4.4 20050721 (Red Hat 3.4.4-2)
g77 -O3 2167.126u 0.235s 36:07.85
g77 -O3 -finit-local-zero -Wno-globals 2127.557u 0.369s 35:29.63
==> 0.0671 CPUs/step in 35 min <-- much slower than zeus-g77
ifort in /opt/intel/fce/9.0/bin/ifort:
Intel(R) Fortran Compiler for Intel(R) EM64T-based v 9.0 Build 20050809
ifort -fast 861.852u 0.153s 14:22.43 2may06
==> 0.0271 CPUs/step in 14 min <-- slower than zeus
newton.usg.utk.edu (head of 36-node Xeon linux cluster)
32 compute nodes: dual Xeon 3.2GHz
uname -a: Linux 2.6.9-11.ELsmp #1 SMP x86_64 x86_64 GNU/Linux
g77 -v: gcc version 3.4.3 20050227 (Red Hat 3.4.3-22.1)
g77 -O3 2328.806u 0.240s 38:50.08
g77 -O3 -finit-local-zero -Wno-globals 2286.761u 0.524s 38:08.21
==> 0.0721 CPUs/step in 38 min <-- slower than zeus
ifort in /opt/intel/fce/9.0/bin/ifort:
Intel(R) Fortran Compiler for Intel(R) EM64T-based v 9.0 Build 20050809
ifort -fast 837.006u 0.132s 13:58.32 2may06
==> 0.0264 CPUs/step in 14 min <-- slower than zeus
ares.math.utk.edu Dell Optiplex 745, Intel Core2 6300
uname -a: Linux 2.6.20-1.2952.fc6 #1 SMP x86_64 GNU/Linux
g77 -v: gcc version 4.1.1 20070105 (Red Hat 4.1.1-51)
g77 -O3 -finit-local-zero 1488.281u 0.281s 24:52.20 3jul07
==> 0.0469 CPUs/step in 25 min
tiger.ornl.gov (head of 144-node Cray XD1 linux cluster)
144 compute nodes: dual Opteron 248
Linux ch328-n6 2.6.5_H_01_04 #39 SMP x86_64 x86_64 GNU/Linux
pgf95 -V: pgf95 7.0-2 64-bit target on x86-64 Linux
pgf95 -fast -O3 -fastsse 836.238u 0.354s 13:57.78
==> 0.0264 CPUs/step in 14 min <-- slower than zeus ?? 2007
g77 -v: gcc version 3.3.3 (SuSE Linux)
g77 -O3 -Wno-globals -funroll-loops 1196.330u 0.457s 19:58.50
==> 0.0377 CPUs/step in 20 min <-- slower than zeus ?? 2007
zeus.math.edu (head of 52-cpu Linux cluster) 5feb2010
head+2 nodes of dual Quad-Core AMD Opteron 2376 2.3GHz 2GB/node
plus 15 dual Opteron 252 nodes 2GB/node
uname -a: head.bw01.math.utk.edu 2.6.18-128.2.1.el5 #1 SMP x86_64
ifort -V: Version 11.1 Build 20090630 ID: l_cprof_p_11.1.046
ifort -fast -O3 563.373u 0.259s 9:23.76
==> 0.017756 CPUs/step in 9 min
midtown.uthsc.edu (head of 56-cpu Linux cluster) 5feb2010
7 nodes of dual Quad-Core AMD Opteron 2376 2.3GHz 2GB/node
uname -a: midtown.bw01.uthsc.edu 2.6.18-164.9.1.el5 #1 SMP x86_64
ifort -V: Version 11.1 Build 20091130 ID: l_cprof_p_11.1.064
ifort -fast -O3 583.247u 0.144s 9:43.65
==> 0.01838 CPUs/step in 10 min
frost.ornl.gov (node of 2048 core Linux cluster) 9mar2010
SGI Altix ICE 8200 cluster, 128 nodes x16=2048, 24GB mem
ifort -fast -O3 424.018u 0.005s 7:04.21
==> 0.01336 CPUs/step in 7:04 min <-- fastest so far
householder.math.utk.edu (20 core cluster) (Fedora19) 26may2014
Two 10 core Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz 192GB mem
uname -a: 3.14.4-100.fc19.x86_64 #1 SMP x86_64 GNU/Linux
gfortran -O3 (gcc 4.8.2) 429.351u 0.006s 7:10.30
==> 0.01353 CPUs/step in 7:10 min <-- a bit slower than frost
mars.math.utk.edu (4 core ) (Fedora21) 26jun2015
Two 2 core Intel(R) Xeon(R) CPU i7-4790 CPU @ 3.60GHz 24.6GB mem
uname -a: 4.0.4-201.fc21.x86_64 #1 SMP x86_64 GNU/Linux
gfortran -O3 (gcc 4.9.2-6) 343.600u 0.011s 5:43.79
==> 0.01083 CPUs/step in 5:44 min <-- fastst so far!!! costs only $1300 !!!
ares.math.utk.edu (4 core ) (Fedora21) 24aug2015
Two 2 core Intel(R) Xeon(R) CPU i7-4790 CPU @ 3.60GHz 16.4GB mem
uname -a: 4.0.8-200.fc21.x86_64 #1 SMP x86_64 GNU/Linux
gfortran -O3 (gcc 4.9.2-6) 345.103u 0.015s 5:45.44
==> 0.01088 CPUs/step in 5:46 min
<-- almost as fast as mars!!! costs only $1300 !!!
Other benchmarking pages: