#!SPARC Station 20 ##Parkbench (LOWLEV_1.0), ##13/SEP/1994, ##at Old Dominion University, Norfolk, Virginia, USA #*name: Roger Hockney #*establishment: University of Southampton #*email: rwh@pac.soton.ac.uk #telephone: 0635 578679 #date sent: 01/11/94 #*date benchmarked: 13/SEP/1994 #*benchmark suite and revision: Parkbench (LOWLEV_1.0) #*message passing library and revision: N/A #*benchmark: RINF1 #*problem size (if applicable): N/A #*number of nodes: 1 #*file number if one of a group: 1 of 1 #*manufacturer: SUN #*model number: SPARC Station 20 #cpu: Sparc #cpu speed (MHz): 50MHz #fpu: integrated #primary cache: #secondary cache: #other cache: #memory: 112MB #disk and connection: #interconnect type: #switching: #other hardware: #operating system and version: Solaris 2.3 #compilers, libraries and versions: f77 version 3.0 #compiler switches: -O2 #additional tuning parameters: none #other software: #additional comments: Benchmarks performed on a workstation # at Old Dominion University, Norfolk, Virginia, USA # in the Center for Coastal Physical Oceanography (CCPO). # There were no other users during the measurements. # ================================================= # === === # === PARKBENCH Parallel Benchmarks === # === === # === RINF1 === # === === # === Program: R-infinity and N-half === # === Version: Standard Fortran 77 === # === Author: Roger Hockney === # === Update: November 1993; Release: 1.0 === # === === # ================================================= # # Run on single SUN SPARC Station 20 # 50MHz without second level cache +112MByte # at Old Dominion University, VA, USA (CCPO) # Software levels are Solaris 2.3 # F77 version 3.0 # Contact: Roger Hockney, rwh@pac.soton.ac.uk # # (1) CONTIGUOUS DYADS: A(I)=B(I)*C(I) # # # ICASE= 1 NCASE= 17 NMAX=100000 NTIMES= 100000 # # #NI TI #vlen sec 1 2.754092E-07 2 4.482031E-07 3 6.003384E-07 4 7.415771E-07 5 9.555221E-07 6 1.069091E-06 7 1.275907E-06 8 1.400757E-06 9 1.565279E-06 10 1.772046E-06 20 1.329827E-05 60 6.502364E-05 70 7.606308E-05 80 1.168566E-04 90 1.364152E-04 100 1.582003E-04 200 3.286409E-04 300 4.888484E-04 400 6.517649E-04 500 8.178055E-04 600 9.738654E-04 700 1.133927E-03 # # SUMMARY OF SELECTED VALUES # -------------------------- # # LENGTHS RMSERR/VALUE R-INFINITY N-HALF R(N) # vlen % Mflop/s vlen Mflop/s # (1) CONTIGUOUS DYADS: A(I)=B(I)*C(I) # <= 90 0.218 6.247 0.712 | Min = 2.739 # >= 700 0.172 2.856 -6.111 | Max = 6.207 # (2) DYADS, STRIDE=8: A(I)=B(I)*C(I) # <= 50 0.787 6.051 0.499 | Min = 0.870 # >= 300 0.231 0.901 -9.380 | Max = 6.090 # # (3) CONTIGUOUS TRIADS: A(I)=B(I)*C(I)+D(I) # <= 90 0.690 9.044 0.630 | Min = 3.635 # >= 700 0.287 3.839 16.883 | Max = 9.029 # # (4) TRIADS, STRIDE=8: A(I)=B(I)*C(I)+D(I) # <= 40 1.503 8.975 0.741 | Min = 1.374 # >= 200 0.093 1.383 -9.735 | Max = 8.869 # # (5) RANDOM SCATTER/GATHER: # <= 400 0.441 6.994 0.507 | Min = 3.627 # >= 2000 0.526 3.763 106.934 | Max = 7.080 # # (6) CONTIGUOUS 4-OP: A(I)=B(I)*C(I)+D(I)*E(I)+F(I) # <= 90 0.293 11.013 0.386 | Min = 5.148 # >= 700 0.126 5.282 14.419 | Max = 11.034 # # (7) INNER PRODUCT: S=S+B(I)*C(I) # <= 300 0.285 11.005 0.411 | Min = 5.632 # >= 1000 0.240 7.214 65.228 | Max = 11.081 # # (8) FIRST ORDER RECURRENCE: A(I)=B(I)*A(I-1)+D(I) # <= 90 0.408 9.074 0.161 | Min = 4.516 # >= 700 1.841 4.939 -156.574 | Max = 14.837 # # (9) CHARGE ASSIGNMENT: A(J(I))=A(J(I))+S # <= 80 0.269 4.970 0.696 | Min = 2.982 # >= 600 0.242 4.371 -19.878 | Max = 4.931 # # (10) TRANSPOSITION: B(I,J)=A(J,I) # <= 10 1.378 12.598 2.792 | Min = 1.982 # >= 80 1.839 1.633 -52.306 | Max = 10.299 # # (11) MATRIX MULT BY INNER PRODUCT # <= 30 1.080 8.911 1.065 | Min = 3.412 # >= 100 2.980 3.437 -46.815 | Max = 8.515 # # (12) MATRIX MULT BY MIDDLE PRODUCT # <= 30 0.818 8.868 0.971 | Min = 3.390 # >= 100 0.272 7.309 2.652 | Max = 8.513 # # (13) MATRIX MULT BY OUTER PRODUCT # <= 64 0.069 1.407 0.486 | Min = 0.984 # >= 60 1.129 0.986 -798.845 | Max = 1.403 # # (14) DYADS, STRIDE=128: A(I)=B(I)*C(I) # <= 8 1.286 6.160 0.711 | Min = 0.609 # >= 60 0.986 0.602 -11.029 | Max = 5.750 # # (15) DYADS, STRIDE=1024: A(I)=B(I)*C(I) # <= 2 0.000 7.289 1.061 | Min = 0.303 # >= 9 3.570 0.274 -6.874 | Max = 5.255 # # (16) CONTIGUOUS DAXPY: A(I)=S*B(I)+C(I) # <= 300 0.181 10.940 0.376 | Min = 5.143 # >= 1000 0.442 5.476 267.866 | Max = 11.026 # # (17) INDIRECT DAXPY: A(J(I))=S*B(K(I))+C(L(I)) # <= 80 0.363 6.591 0.370 | Min = 1.866 # >= 600 0.063 1.929 16.315 | Max = 6.592 # # # TOTAL EXECUTION TIME IS 1.1017758179E+02 SECONDS.