#!Indigo 2 XL ##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: Silicon Graphics Incorporated #*model number: Indigo 2 XL #cpu: R4400 #cpu speed (MHz): 150MHz #fpu: integrated #primary cache: #secondary cache: #other cache: #memory: #disk and connection: #interconnect type: #switching: #other hardware: #operating system and version: IRIX Rel. 5.2 IP22 #compilers, libraries and versions: f77 #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 Silicon Graphics SGI Indigo 2 XL # single R4400 CPU at 150MHz # at Old Dominion University, VA, USA (CCPO) # Running under IRIX release 5.2 IP22 # F77 version # 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 2.706170E-07 2 3.040314E-07 3 4.806284E-07 4 1.199722E-07 5 7.210970E-07 6 9.255535E-07 10 1.308083E-06 20 4.021883E-06 30 8.943202E-06 40 1.351595E-05 50 1.679480E-05 60 1.856649E-05 70 2.168350E-05 80 2.517605E-05 90 3.097401E-05 100 3.232956E-05 200 7.938147E-05 300 1.292120E-04 400 1.639986E-04 500 2.212524E-04 600 2.436265E-04 700 3.127882E-04 # # 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) # <= 100 5.231 14.312 3.961 | Min = 2.590 # >= 800 4.283 2.735 -3354.627 | Max = 14.247 # # (2) DYADS, STRIDE=8: A(I)=B(I)*C(I) # <= 30 2.281 8.532 1.619 | Min = 0.604 # >= 100 7.569 0.634 -768.013 | Max = 8.127 # # (3) CONTIGUOUS TRIADS: A(I)=B(I)*C(I)+D(I) # <= 100 6.815 20.355 1.871 | Min = 3.685 # >= 800 4.479 3.833 -3716.703 | Max = 21.786 # # (4) TRIADS, STRIDE=8: A(I)=B(I)*C(I)+D(I) # <= 20 6.870 12.622 0.431 | Min = 0.727 # >= 90 8.444 0.784 -769.286 | Max = 34.121 # # (5) RANDOM SCATTER/GATHER: # <= 300 0.687 15.368 2.999 | Min = 4.432 # >= 1000 2.124 4.677 -1140.707 | Max = 15.245 # # (6) CONTIGUOUS 4-OP: A(I)=B(I)*C(I)+D(I)*E(I)+F(I) # <= 100 3.675 27.074 1.963 | Min = 5.038 # >= 800 1.881 5.034 -3107.607 | Max = 28.777 # # (7) INNER PRODUCT: S=S+B(I)*C(I) # <= 200 4.066 31.093 4.769 | Min = 6.616 # >= 900 2.224 9.345 -2016.292 | Max = 31.495 # # (8) FIRST ORDER RECURRENCE: A(I)=B(I)*A(I-1)+D(I) # <= 100 2.846 20.336 1.729 | Min = 4.951 # >= 800 3.884 5.176 -3300.406 | Max = 31.841 # # (9) CHARGE ASSIGNMENT: A(J(I))=A(J(I))+S # <= 80 3.481 10.963 2.381 | Min = 3.200 # >= 600 0.925 6.157 343.460 | Max = 11.068 # # (10) TRANSPOSITION: B(I,J)=A(J,I) # <= 9 8.190 45.414 20.616 | Min = 2.682 # >= 70 2.768 3.028 -40.460 | Max = 14.990 # # (11) MATRIX MULT BY INNER PRODUCT # <= 80 2.631 12.246 0.649 | Min = 3.408 # >= 600 0.000 9.203 -5.483 | Max = 13.713 # # (12) MATRIX MULT BY MIDDLE PRODUCT # <= 60 3.324 18.451 2.511 | Min = 3.573 # >= 400 0.000 15.168 -0.126 | Max = 19.790 # # (13) MATRIX MULT BY OUTER PRODUCT # <= 81 5.433 10.281 6.208 | Min = 2.519 # >= 70 4.214 2.961 -4808.752 | Max = 10.371 # # (14) DYADS, STRIDE=128: A(I)=B(I)*C(I) # <= 3 7.022 9.523 1.350 | Min = 2.238 # >= 10 1.796 2.263 -15.937 | Max = 33.341 # # (15) DYADS, STRIDE=1024: A(I)=B(I)*C(I) # <= 1 0.000 0.000 0.000 | Min = 1.066 # >= 8 0.619 0.998 -6.271 | Max = 5.491 # # (16) CONTIGUOUS DAXPY: A(I)=S*B(I)+C(I) # <= 200 1.864 23.314 3.706 | Min = 4.331 # >= 900 4.062 5.344 -3088.961 | Max = 34.438 # # (17) INDIRECT DAXPY: A(J(I))=S*B(K(I))+C(L(I)) # <= 80 3.503 11.287 2.019 | Min = 1.735 # >= 600 1.559 1.774 -1881.928 | Max = 11.501 # # # # TOTAL EXECUTION TIME IS 3.0220371246E+01 SECONDS.