Hi all
 
I'm trying to benchmark my cluster using HPL (linpack). I can run HPL fine and it gives good results running only on the master node. However whentrying to run across the cluster, the process hangs. It goes to a certain point :
 
- The relative machine precision (eps) is taken to be     1.11022e-16
- Computational tests pass if scaled residuals are less than    16
 
 
 
the process then just stops there and nothing else happens.
 
I'm using MPICH as my default MPI implementation.
 
The command given to execute xhpl on 2 nodes is :
 
> mpirun -np 2 -machinefile /home/machines.LINUX xhpl
 
My HPL.dat file looks like this :
 
HPLinpack benchmark input file
Innovative Computing Laboratory, University of Tennessee
HPL.out      output file name (if any)
6            device out (6=stdout,7=stderr,file)
1            # of problems sizes (N)
1000 Ns
1            # of NBs
64 NBs
1            # of process grids (P x Q)
1 Ps
2 Qs
16.0         threshold
3            # of panel fact
0 1 2            PFACTs (0=left, 1=Crout, 2=Right)
1            # of recursive stopping criterium
8            NBMINs (>= 1)
1            # of panels in recursion
2            NDIVs
1            # of recursive panel fact.
2            RFACTs (0=left, 1=Crout, 2=Right)
1            # of broadcast
1            BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)
1            # of lookahead depth
1            DEPTHs (>=0)
2            SWAP (0=bin-exch,1=long,2=mix)
80           swapping threshold
0            L1 in (0=transposed,1=no-transposed) form
0            U  in (0=transposed,1=no-transposed) form
1            Equilibration (0=no,1=yes)
8            memory alignment in double (> 0)
 
 
 
 
Does anyone have a clue as to what i'm doing wrong? Has anyone benchmarked their OSCAR clusters using HPL?
 
Your help would be greatly appreciated
 
Regards
Ralph

Reply via email to