xdlutime results

[Putchong Uthayopas]

Hi,
MFlops means Millions Floatpoint Operation Per Seconds. It is the measures
of how fast you systems can do floating point operations. It is a general
performance indication
for scientific computer.

Below are our results. The machines is 8 Athlon 550MHz and one 1Ghz Athlon.
512 Mb memory each. and Myrinet.
More detail of the system can be found at
http://smile.cpe.ku.ac.th/cluster/athlon.html

Putchong
Kasetsart University, Thailand.
========================================================

webadmin at amata84 TESTING]$ mpirun -np 9 xdlutime LUTIME.dat
Simple Timer for ScaLAPACK routine PDGESV

Number of processors used: 9

TIME N NB P Q LU Time Sol Time MFLOP/S Residual CHECK

---- ----- --- --- --- --------- --------- -------- -------- -------

WALL 6000 9 1 9 116.80 0.41 1229.04 0.001097 PASSED

WALL 6000 9 9 1 200.79 3.65 704.66 0.001926 PASSED

Comments for the output:

Column 1: the time is the wall clock time

Column 2 & 3: matrix size (N by N) and block size (NB by NB)

Column 4 & 5: grid size (P by Q)

Column 6: total time (in seconds) to run PDGESV

Column 7: mega flops per seconds

Column 8: if PDGESV works fine

----- Original Message -----
From: "Horatio B. Bogbindero" <wyy at cersa.admu.edu.ph>
To: <beowulf at beowulf.org>
Sent: Wednesday, September 13, 2000 12:35 PM
Subject: xdlutime results


>
> i need help in interpreting my xdlutime results in a 8-node beowulf
> cluster connected via 100mbps fast ethernet (intel 410T and
> inteleepro100s).
>
> here are the results:
> --------------------
>
> Simple Timer for ScaLAPACK routine PDGESV
> Number of processors used:   8
>
> TIME     N  NB   P   Q  LU Time   Sol Time  MFLOP/S Residual  CHECK
> ---- ----- --- --- --- --------- --------- -------- -------- -------
> WALL  2828  64   2   2     42.70      0.35   350.49 0.005429 PASSED
> WALL  2828  64   2   2     42.69      0.35   350.60 0.005429 PASSED
>
> --------------------
>
> what does the MFLOP/S here mean? thanks.
>
> ---------------------
> william.s.yu at ieee.org
>
> Good news from afar can bring you a welcome visitor.
>
>
>
> _______________________________________________
> Beowulf mailing list
> Beowulf at beowulf.org
> http://www.beowulf.org/mailman/listinfo/beowulf