Seattle: Building and running HPL - Revision history

Jon: /* Results */

2015-01-09T18:52:41Z

Results

2015-01-09T16:38:14Z

Results

2015-01-09T15:21:40Z

Results

2015-01-09T14:17:02Z

Results

2015-01-09T14:15:38Z

Results

2015-01-09T12:56:36Z

Edit the HPL.dat to optimise

2015-01-09T12:53:33Z

Download & Build Linpack

2015-01-09T12:53:00Z

Install OpenMPI

2015-01-09T12:48:00Z

Results

2015-01-09T12:47:32Z

Results

@@ Line 144: / Line 144: @@
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 384 || 2x3 || Overdrive-demo || 9.809
 |-
-| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 43200 || 384 || 2x3 || Overdrive-demo || tbc
+| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 43200 || 384 || 2x3 || Overdrive-demo || 9.832
 |}

@@ Line 142: / Line 142: @@
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 256 || 2x3 || Overdrive-demo || 9.738
 |-
-| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 384 || 2x3 || Overdrive-demo || tbc
+| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 384 || 2x3 || Overdrive-demo || 9.809
 |}

@@ Line 140: / Line 140: @@
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 128 || 2x3 || Overdrive-demo || 9.054
 |-
-| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 256 || 2x3 || Overdrive-demo || tbc
+| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 256 || 2x3 || Overdrive-demo || 9.738
 |}

@@ Line 139: / Line 139: @@
 |-
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 128 || 2x3 || Overdrive-demo || 9.054
 |}

@@ Line 138: / Line 138: @@
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 35328 || 128 || 2x3 || Overdrive-demo || 8.963
 |-
-| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 128 || 2x3 || Overdrive-demo || tbc
+| scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 38010 || 128 || 2x3 || Overdrive-demo || 9.054
 |}

@@ Line 94: / Line 94: @@
 * '''Block Size (NB):''' HPL uses the block size NB for the data distribution as well as for the computational granularity. A very small NB will limit computational performance because no data reuse will occur, and also the number of messages will also increase. "Good" block sizes are almost always in the [32 .. 256] interval and it  depends on  Cache size.  These block size are found to be good, 80-216 for IA32; 128-192 for IA64 3M cache; 400 for 4M cache for IA64 and 130 for Woodcrests.
-* '''Process Grid Ratio (PXQ):''' This depends on physical interconnection network. P and Q should be approximately equal, with Q slightly larger than P. P * Q should equal the number of available cores. Our sample system had a 6-core CPU to '''P=2''' & '''Q=3'''
+* '''Process Grid Ratio (PXQ):''' This depends on physical interconnection network. P and Q should be approximately equal, with Q slightly larger than P. P * Q should equal the number of available cores. Our sample system had a 6-core CPU, so '''P = 2''' & '''Q = 3'''
 These numbers can be auto generated by a number of online tools, for example: http://www.advancedclustering.com/act-kb/tune-hpl-dat-file/

@@ Line 73: / Line 73: @@
 LAdir        = /root/scratch/atlas3.10.1_build/lib
 ---
-LINKER       = /usr/lib64/openmpi/bin/mpif90
+LINKER      = /usr/lib64/openmpi/bin/mpif90
 </syntaxhighlight>

@@ Line 8: / Line 8: @@
 == Install OpenMPI ==
-Either install from the repos '''<code>openmpi</code>''', build it manually of choose an alternative MPI.
+Either install from the repos '''<code>openmpi</code>''', build a copy manually or choose an alternative MPI.
 <syntaxhighlight>
 yum install openmpi openmpi-devel

@@ Line 130: / Line 130: @@
 == Results ==
 {| class="wikitable" style="text-align:center; width:100%; "
-|+ Intel Linpack results for Dual CPU system
+|+ Intel Linpack results for Single SoC system
 |-
 ! colspan="11" | Haswell

@@ Line 134: / Line 134: @@
 ! colspan="11" | Haswell
 |-
-! CPU || Freq || Cores || Memory (GB) || N || NB || PxQ || System || Result
+! CPU || Freq || Cores || Memory (GB) || N || NB || PxQ || System || Result (GFLOPS)
 |-
 | scope="row" | AMD Seattle || 1.50GHz || 6 || 16 || 35328 || 128 || 2x3 || Overdrive-demo || 8.963
 |}