🚀 go-pugleaf

In article <27SEP92.14332660@wl.aecl.ca>, harrisp@wl.aecl.ca writes:
>  We are considering purchasing a VS-4000-90. However, I've got very little
> in the way of performance benchmarks for it. I'd really be interested in
> knowing if anyone out there has run the LINPACK benchmark, and what numbers
> they got out. Specmarks are nice, but they never seem to correlate well with
> the performance I see when I use the box, but the LINPACK measurements do.
> Maybe that's cause I use a lot of the Linpack codes. (sorry for rambling
> a bit here) Alternatively, if some kind soul would be willing to run the
> Linpack benchmark on there system for me, I'd be more than happy to mail
> them the code.
> 
>                 Thanks in advance
>                                 Phil
Here are the results of the Fortran Linpack benchmark (as posted by
eoq@vthnw.cvm.msu.edu (Edward Quillen) on 6 Sep 92 in comp.benchmarks) on both
a VAXstation 3100-76 and 4000-90:

VAXstation 3100-76:
*******************

<IRT>run linpack
Please send the results of this run to:

Jack J. Dongarra
Mathematics and Computer Science Division
Argonne National Laboratory
Argonne, Illinois 60439

Telephone: 312-972-7246

ARPAnet: DONGARRA@ANL-MCS

    norm. resid      resid           machep         x(1)          x(n)
 2.10272591E+00  1.16642807E-14  2.77555756E-17  1.00000000E+00  1.00000000E+00

   times are reported for matrices of order   100
     dgefa      dgesl      total     mflops       unit      ratio
times for array with leading dimension of 201
 7.930E-01  2.734E-02  8.203E-01  8.371E-01  2.389E+00  1.465E+01
 7.695E-01  5.078E-02  8.203E-01  8.371E-01  2.389E+00  1.465E+01
 7.813E-01  3.125E-02  8.125E-01  8.451E-01  2.367E+00  1.451E+01
 7.789E-01  2.891E-02  8.078E-01  8.500E-01  2.353E+00  1.443E+01

times for array with leading dimension of 200
 8.086E-01  3.125E-02  8.398E-01  8.176E-01  2.446E+00  1.500E+01
 7.891E-01  4.297E-02  8.320E-01  8.253E-01  2.423E+00  1.486E+01
 7.813E-01  1.953E-02  8.008E-01  8.575E-01  2.332E+00  1.430E+01
 7.898E-01  2.773E-02  8.176E-01  8.399E-01  2.381E+00  1.460E+01
FORTRAN STOP

VAXstation 4000-90
******************

$ run linpack
Please send the results of this run to:

Jack J. Dongarra
Mathematics and Computer Science Division
Argonne National Laboratory
Argonne, Illinois 60439

Telephone: 312-972-7246

ARPAnet: DONGARRA@ANL-MCS

    norm. resid      resid           machep         x(1)          x(n)
 2.10272591E+00  1.16642807E-14  2.77555756E-17  1.00000000E+00  1.00000000E+00


   times are reported for matrices of order   100
     dgefa      dgesl      total     mflops       unit      ratio
times for array with leading dimension of 201
 1.289E-01  0.000E+00  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.289E-01  0.000E+00  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.211E-01  7.813E-03  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.305E-01  3.906E-03  1.344E-01  5.110E+00  3.914E-01  2.400E+00

times for array with leading dimension of 200
 1.211E-01  7.813E-03  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.289E-01  0.000E+00  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.172E-01  1.172E-02  1.289E-01  5.327E+00  3.755E-01  2.302E+00
 1.246E-01  4.297E-03  1.289E-01  5.327E+00  3.755E-01  2.302E+00
FORTRAN STOP


-- 
Jean-Pierre Farine
------------------------------------------------------------------------------
e-mail:		farine@decus.ch
x.400:		c=ch,admd=arcom,o=grd,s=farine
FAX:		+41 31 67 61 21		Phone:	+41 31 67 61 10 / 13
------------------------------------------------------------------------------