This is a version of the Nasa Parallel Benchmark prepared by Doug Eadline of Paralogic (a turnkey beowulf vendor). The question is whether the tasks running one at a time on a dual complete in half the time when split up in parallel on a dual. These are "real" parallel tasks, run over MPI-LAM, and so this is a true "beowulf" benchmark.<\p>
Wed Mar 14 15:32:55 PST 2001 ================== PROGRAM CG ================== NAS Parallel Benchmarks 2.3 -- CG Benchmark Size: 14000 Iterations: 15 Number of active processes: 1 iteration ||r|| zeta 1 0.27830335051741E-12 19.9997581277040 2 0.26267465793923E-14 17.1140495745506 3 0.26593258594959E-14 17.1296668946143 4 0.26027437659444E-14 17.1302113581193 5 0.26415698292524E-14 17.1302338856353 6 0.26593580261729E-14 17.1302349879482 7 0.26522430119504E-14 17.1302350498916 8 0.25848809208742E-14 17.1302350537510 9 0.25682718194417E-14 17.1302350540101 10 0.25518224955758E-14 17.1302350540284 11 0.25299808618898E-14 17.1302350540298 12 0.25443530675953E-14 17.1302350540299 13 0.25865405354399E-14 17.1302350540299 14 0.25427678325957E-14 17.1302350540299 15 0.25368139570097E-14 17.1302350540299 Benchmark completed VERIFICATION SUCCESSFUL Zeta is 0.171302350540E+02 Error is 0.877520278664E-12 CG Benchmark Completed. Class = A Size = 14000 Iterations = 15 Time in seconds = 36.60 Total processes = 1 Compiled procs = 1 Mop/s total = 40.89 Mop/s/process = 40.89 Operation type = floating point Verification = SUCCESSFUL Version = 2.3 Compile date = 20 Feb 2001 Compile options: MPIF77 = f77 FLINK = f77 FMPI_LIB = -L../MPI_dummy -lmpi FMPI_INC = -I../MPI_dummy FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = randi8 Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM EP ================== NAS Parallel Benchmarks 2.3 -- EP Benchmark Number of random numbers generated: 536870912 Number of active processes: 1 EP Benchmark Results: CPU Time = 136.5274 N = 2^ 28 No. Gaussian Pairs = 210832767. Sums = -4.295875165629841E+03 -1.580732573678432E+04 Counts: 0 98257395. 1 93827014. 2 17611549. 3 1110028. 4 26536. 5 245. 6 0. 7 0. 8 0. 9 0. EP Benchmark Completed. Class = A Size = 536870912 Iterations = 0 Time in seconds = 136.53 Total processes = 1 Compiled procs = 1 Mop/s total = 3.93 Mop/s/process = 3.93 Operation type = Random numbers generated Verification = SUCCESSFUL Version = 2.3 Compile date = 20 Feb 2001 Compile options: MPIF77 = f77 FLINK = f77 FMPI_LIB = -L../MPI_dummy -lmpi FMPI_INC = -I../MPI_dummy FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = randi8 Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM IS ================== NAS Parallel Benchmarks 2.3 -- IS Benchmark Size: 8388608 (class A) Iterations: 10 Number of processes: 1 iteration 1 2 3 4 5 6 7 8 9 10 IS Benchmark Completed Class = A Size = 8388608 Iterations = 10 Time in seconds = 9.70 Total processes = 1 Compiled procs = 1 Mop/s total = 8.65 Mop/s/process = 8.65 Operation type = keys ranked Verification = SUCCESSFUL Version = 2.3 Compile date = 20 Feb 2001 Compile options: MPICC = cc CLINK = cc CMPI_LIB = -L../MPI_dummy -lmpi CMPI_INC = -I../MPI_dummy CFLAGS = -O2 -malign-double -ffast-math CLINKFLAGS = (none) Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM LU ================== NAS Parallel Benchmarks 2.2 -- LU Benchmark Size: 64x 64x 64 Iterations: 250 Number of processes: 1 Time step 1 Time step 20 Time step 40 Time step 60 Time step 80 Time step 100 Time step 120 Time step 140 Time step 160 Time step 180 Time step 200 Time step 220 Time step 240 Time step 250 Verification being performed for class A Accuracy setting for epsilon = 0.1000000000000E-07 Comparison of RMS-norms of residual 1 0.7790210760669E+03 0.7790210760669E+03 0.8756130575742E-15 2 0.6340276525969E+02 0.6340276525969E+02 0.1008612888673E-14 3 0.1949924972729E+03 0.1949924972729E+03 0.7287898208366E-15 4 0.1784530116042E+03 0.1784530116042E+03 0.3026076580738E-14 5 0.1838476034946E+04 0.1838476034946E+04 0.2102476403859E-14 Comparison of RMS-norms of solution error 1 0.2996408568547E+02 0.2996408568547E+02 0.8299601066640E-15 2 0.2819457636500E+01 0.2819457636500E+01 0.7875436823397E-15 3 0.7347341269877E+01 0.7347341269877E+01 0.4835373161943E-15 4 0.6713922568778E+01 0.6713922568778E+01 0.5291561888607E-15 5 0.7071531568839E+02 0.7071531568839E+02 0.6028759644299E-15 Comparison of surface integral 0.2603092560489E+02 0.2603092560489E+02 0.4094414927144E-15 Verification Successful LU Benchmark Completed. Class = A Size = 64x 64x 64 Iterations = 250 Time in seconds = 812.79 Total processes = 1 Compiled procs = 1 Mop/s total = 146.78 Mop/s/process = 146.78 Operation type = floating point Verification = SUCCESSFUL Version = 2.3 Compile date = 20 Feb 2001 Compile options: MPIF77 = f77 FLINK = f77 FMPI_LIB = -L../MPI_dummy -lmpi FMPI_INC = -I../MPI_dummy FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = (none) Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 Wed Mar 14 15:49:44 PST 2001 <\pre>
Dual Results
Wed Mar 14 16:04:24 PST 2001 LAM 6.4-a3/IMPI/MPI 2 C++ - University of Notre Dame Executing hboot on n0 (205-219-89-197)... topology n0 (o)... done ================== PROGRAM CG ================== NAS Parallel Benchmarks 2.3 -- CG Benchmark Size: 14000 Iterations: 15 Number of active processes: 2 iteration ||r|| zeta 1 0.30291810122364E-12 19.9997581277040 2 0.30430868884537E-14 17.1140495745506 3 0.30424122448075E-14 17.1296668946143 4 0.30809127937477E-14 17.1302113581192 5 0.30430378474328E-14 17.1302338856353 6 0.30295060108641E-14 17.1302349879482 7 0.29844192331483E-14 17.1302350498916 8 0.29687013833879E-14 17.1302350537510 9 0.30647175210656E-14 17.1302350540101 10 0.29778179159041E-14 17.1302350540284 11 0.30412965175183E-14 17.1302350540298 12 0.30104552653157E-14 17.1302350540299 13 0.29606125488682E-14 17.1302350540299 14 0.30380073566568E-14 17.1302350540299 15 0.29871187919337E-14 17.1302350540299 Benchmark completed VERIFICATION SUCCESSFUL Zeta is 0.171302350540E+02 Error is 0.895283847058E-12 CG Benchmark Completed. Class = A Size = 14000 Iterations = 15 Time in seconds = 19.17 Total processes = 2 Compiled procs = 2 Mop/s total = 78.06 Mop/s/process = 39.03 Operation type = floating point Verification = SUCCESSFUL Version = 2.3 Compile date = 18 Dec 2000 Compile options: MPIF77 = hf77 FLINK = hf77 FMPI_LIB = -L/usr/local/lib -lmpi FMPI_INC = -I/usr/local/lam62/h FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = randi8 Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM EP ================== NAS Parallel Benchmarks 2.3 -- EP Benchmark Number of random numbers generated: 536870912 Number of active processes: 2 EP Benchmark Results: CPU Time = 67.8625 N = 2^ 28 No. Gaussian Pairs = 210832767. Sums = -4.295875165639063E+03 -1.580732573678573E+04 Counts: 0 98257395. 1 93827014. 2 17611549. 3 1110028. 4 26536. 5 245. 6 0. 7 0. 8 0. 9 0. EP Benchmark Completed. Class = A Size = 536870912 Iterations = 0 Time in seconds = 67.86 Total processes = 2 Compiled procs = 2 Mop/s total = 7.91 Mop/s/process = 3.96 Operation type = Random numbers generated Verification = SUCCESSFUL Version = 2.3 Compile date = 18 Dec 2000 Compile options: MPIF77 = hf77 FLINK = hf77 FMPI_LIB = -L/usr/local/lib -lmpi FMPI_INC = -I/usr/local/lam62/h FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = randi8 Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM IS ================== NAS Parallel Benchmarks 2.3 -- IS Benchmark Size: 8388608 (class A) Iterations: 10 Number of processes: 2 iteration 1 2 3 4 5 6 7 8 9 10 IS Benchmark Completed Class = A Size = 8388608 Iterations = 10 Time in seconds = 4.83 Total processes = 2 Compiled procs = 2 Mop/s total = 17.35 Mop/s/process = 8.68 Operation type = keys ranked Verification = SUCCESSFUL Version = 2.3 Compile date = 18 Dec 2000 Compile options: MPICC = hcc CLINK = hcc CMPI_LIB = -L/usr/local/lib -lmpi CMPI_INC = -I/usr/local/lam62/h CFLAGS = -O2 -malign-double -ffast-math CLINKFLAGS = (none) Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 ================== PROGRAM LU ================== NAS Parallel Benchmarks 2.2 -- LU Benchmark Size: 64x 64x 64 Iterations: 250 Number of processes: 2 Time step 1 Time step 20 Time step 40 Time step 60 Time step 80 Time step 100 Time step 120 Time step 140 Time step 160 Time step 180 Time step 200 Time step 220 Time step 240 Time step 250 Verification being performed for class A Accuracy setting for epsilon = 0.1000000000000E-07 Comparison of RMS-norms of residual 1 0.7790210760669E+03 0.7790210760669E+03 0.1795006768027E-13 2 0.6340276525969E+02 0.6340276525969E+02 0.8405107405605E-14 3 0.1949924972729E+03 0.1949924972729E+03 0.5684560602526E-14 4 0.1784530116042E+03 0.1784530116042E+03 0.1592671884599E-15 5 0.1838476034946E+04 0.1838476034946E+04 0.1644878598313E-13 Comparison of RMS-norms of solution error 1 0.2996408568547E+02 0.2996408568547E+02 0.2015617401898E-14 2 0.2819457636500E+01 0.2819457636500E+01 0.2047613574083E-14 3 0.7347341269878E+01 0.7347341269877E+01 0.6527753768624E-14 4 0.6713922568778E+01 0.6713922568778E+01 0.5423850935823E-14 5 0.7071531568839E+02 0.7071531568839E+02 0.9846974085689E-14 Comparison of surface integral 0.2603092560489E+02 0.2603092560489E+02 0.0000000000000E+00 Verification Successful LU Benchmark Completed. Class = A Size = 64x 64x 64 Iterations = 250 Time in seconds = 432.50 Total processes = 2 Compiled procs = 2 Mop/s total = 275.83 Mop/s/process = 137.92 Operation type = floating point Verification = SUCCESSFUL Version = 2.3 Compile date = 18 Dec 2000 Compile options: MPIF77 = hf77 FLINK = hf77 FMPI_LIB = -L/usr/local/lib -lmpi FMPI_INC = -I/usr/local/lam62/h FFLAGS = -O2 -malign-double -ffast-math FLINKFLAGS = (none) RAND = (none) Please send the results of this run to: NPB Development Team Internet: npb@nas.nasa.gov If email is not available, send this to: MS T27A-1 NASA Ames Research Center Moffett Field, CA 94035-1000 Fax: 415-604-3957 Wed Mar 14 16:13:18 PST 2001 <\pre>