Difference between revisions of "HPL.out"

From In The Wings
Jump to navigation Jump to search
 
(One intermediate revision by the same user not shown)
Line 3: Line 3:
  
 
<pre>
 
<pre>
 +
 
==Second Attempt==
 
==Second Attempt==
 +
 
<pre>
 
<pre>
−
============================================================================
 
−
HPLinpack 1.0a  --  High-Performance Linpack benchmark  --  January 20, 2004
 
−
Written by A. Petitet and R. Clint Whaley,  Innovative Computing Labs.,  UTK
 
−
============================================================================
 
−
 
−
An explanation of the input/output parameters follows:
 
−
T/V    : Wall time / encoded variant.
 
−
N      : The order of the coefficient matrix A.
 
−
NB    : The partitioning blocking factor.
 
−
P      : The number of process rows.
 
−
Q      : The number of process columns.
 
−
Time  : Time in seconds to solve the linear system.
 
−
Gflops : Rate of execution for solving the linear system.
 
−
 
−
The following parameter values will be used:
 
−
 
−
N      :    8192
 
−
NB    :      16
 
−
PMAP  : Row-major process mapping
 
−
P      :      1
 
−
Q      :      1
 
−
PFACT  :    Left    Crout    Right
 
−
NBMIN  :      2        4
 
−
NDIV  :      2
 
−
RFACT  :    Left    Crout    Right
 
−
BCAST  :  1ring
 
−
DEPTH  :      0
 
−
SWAP  : Mix (threshold = 64)
 
−
L1    : transposed form
 
−
U      : transposed form
 
−
EQUIL  : yes
 
−
ALIGN  : 8 double precision words
 
−
 
−
----------------------------------------------------------------------------
 
−
 
−
- The matrix A is randomly generated for each test.
 
−
- The following scaled residual checks will be computed:
 
−
  1) ||Ax-b||_oo / ( eps * ||A||_1  * N        )
 
−
  2) ||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  )
 
−
  3) ||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo )
 
−
- The relative machine precision (eps) is taken to be          1.110223e-16
 
−
- Computational tests pass if scaled residuals are less than          16.0
 
  
 
============================================================================
 
============================================================================
Line 65: Line 25:
 
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED
 
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED
 
</pre>
 
</pre>
 +
 
==Third Attempt==
 
==Third Attempt==
 +
 
<pre>
 
<pre>
 +
 
============================================================================
 
============================================================================
 
T/V                N    NB    P    Q              Time            Gflops
 
T/V                N    NB    P    Q              Time            Gflops
Line 84: Line 47:
 
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0041571 ...... PASSED
 
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0041571 ...... PASSED
 
</pre>
 
</pre>
−
==Fourth Attempt==
 
−
 
−
============================================================================
 
−
 
−
An explanation of the input/output parameters follows:
 
−
T/V    : Wall time / encoded variant.
 
−
N      : The order of the coefficient matrix A.
 
−
NB    : The partitioning blocking factor.
 
−
P      : The number of process rows.
 
−
Q      : The number of process columns.
 
−
Time  : Time in seconds to solve the linear system.
 
−
Gflops : Rate of execution for solving the linear system.
 
−
 
−
The following parameter values will be used:
 
−
 
−
N      :    8192
 
−
NB    :      16
 
−
PMAP  : Row-major process mapping
 
−
P      :      1
 
−
Q      :      1
 
−
PFACT  :    Left    Crout    Right
 
−
NBMIN  :      2        4
 
−
NDIV  :      2
 
−
RFACT  :    Left    Crout    Right
 
−
BCAST  :  1ring
 
−
DEPTH  :      0
 
−
SWAP  : Mix (threshold = 64)
 
−
L1    : transposed form
 
−
U      : transposed form
 
−
EQUIL  : yes
 
−
ALIGN  : 8 double precision words
 
−
 
−
----------------------------------------------------------------------------
 
  
−
- The matrix A is randomly generated for each test.
 
−
- The following scaled residual checks will be computed:
 
−
  1) ||Ax-b||_oo / ( eps * ||A||_1  * N        )
 
−
  2) ||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  )
 
−
  3) ||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo )
 
−
- The relative machine precision (eps) is taken to be          1.110223e-16
 
−
- Computational tests pass if scaled residuals are less than          16.0
 
−
 
−
============================================================================
 
−
T/V                N    NB    P    Q              Time            Gflops
 
−
----------------------------------------------------------------------------
 
−
WR00L2L2        8192    16    1    1            8794.08          4.169e-02
 
−
----------------------------------------------------------------------------
 
−
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0522414 ...... PASSED
 
−
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0220388 ...... PASSED
 
−
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0036372 ...... PASSED
 
 
============================================================================
 
============================================================================
 
T/V                N    NB    P    Q              Time            Gflops
 
T/V                N    NB    P    Q              Time            Gflops
Line 144: Line 58:
  
 
</pre>
 
</pre>
 +
 
==Second Attempt==
 
==Second Attempt==
 +
 
<pre>
 
<pre>
−
============================================================================
 
−
HPLinpack 1.0a  --  High-Performance Linpack benchmark  --  January 20, 2004
 
−
Written by A. Petitet and R. Clint Whaley,  Innovative Computing Labs.,  UTK
 
−
============================================================================
 
−
 
−
An explanation of the input/output parameters follows:
 
−
T/V    : Wall time / encoded variant.
 
−
N      : The order of the coefficient matrix A.
 
−
NB    : The partitioning blocking factor.
 
−
P      : The number of process rows.
 
−
Q      : The number of process columns.
 
−
Time  : Time in seconds to solve the linear system.
 
−
Gflops : Rate of execution for solving the linear system.
 
−
 
−
The following parameter values will be used:
 
−
 
−
N      :    8192
 
−
NB    :      16
 
−
PMAP  : Row-major process mapping
 
−
P      :      1
 
−
Q      :      1
 
−
PFACT  :    Left    Crout    Right
 
−
NBMIN  :      2        4
 
−
NDIV  :      2
 
−
RFACT  :    Left    Crout    Right
 
−
BCAST  :  1ring
 
−
DEPTH  :      0
 
−
SWAP  : Mix (threshold = 64)
 
−
L1    : transposed form
 
−
U      : transposed form
 
−
EQUIL  : yes
 
−
ALIGN  : 8 double precision words
 
−
 
−
----------------------------------------------------------------------------
 
−
 
−
- The matrix A is randomly generated for each test.
 
−
- The following scaled residual checks will be computed:
 
−
  1) ||Ax-b||_oo / ( eps * ||A||_1  * N        )
 
−
  2) ||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  )
 
−
  3) ||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo )
 
−
- The relative machine precision (eps) is taken to be          1.110223e-16
 
−
- Computational tests pass if scaled residuals are less than          16.0
 
  
 
============================================================================
 
============================================================================

Latest revision as of 11:21, 28 August 2007

First attempt

This attempt is the first run I did. It probably needs tuning because it SUCKS! This is the result I received with the use of VSIPL as the math libraries. I am now working on getting BLAS installed to test against those instead. 


==Second Attempt==

<pre>

============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L2        8192    16     1     1            1554.50          2.358e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0625347 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0263812 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED
============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L4        8192    16     1     1            1546.80          2.370e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0625347 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0263812 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED

Third Attempt


============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L2        8192    32     1     1            1203.06          3.047e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0597090 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0251891 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0041571 ...... PASSED
============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L4        8192    32     1     1            1202.94          3.048e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0597090 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0251891 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0041571 ...... PASSED
================================================================

T/V N NB P Q Time Gflops

WR00L2L4 8192 16 1 1 8676.29 4.225e-02

||Ax-b||_oo / ( eps * ||A||_1 * N ) = 0.0522414 ...... PASSED ||Ax-b||_oo / ( eps * ||A||_1 * ||x||_1 ) = 0.0220388 ...... PASSED ||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) = 0.0036372 ...... PASSED

Second Attempt


============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L2        8192    16     1     1            1554.50          2.358e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0625347 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0263812 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED
============================================================================
T/V                N    NB     P     Q               Time             Gflops
----------------------------------------------------------------------------
WR00L2L4        8192    16     1     1            1546.80          2.370e-01
----------------------------------------------------------------------------
||Ax-b||_oo / ( eps * ||A||_1  * N        ) =        0.0625347 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_1  * ||x||_1  ) =        0.0263812 ...... PASSED
||Ax-b||_oo / ( eps * ||A||_oo * ||x||_oo ) =        0.0043538 ...... PASSED
Retrieved from "https://wiki.inthewings.net/index.php?title=HPL.out&oldid=2239"