Differences

This shows you the differences between two versions of the page.

--- public:benchmark [2015/01/14 15:28] – [LALInference Nest] paul.hopkins@LIGO.ORG
+++ public:benchmark [2015/03/18 23:58] (current) – [LALInference Nest] paul.hopkins@LIGO.ORG
@@ Line 338: / Line 338: @@
 Using an NVidia Tesla K10 this completed in a time of 330 seconds.
-===== Build LALSuite (Again) =====
-Previous build instructions minimal set, this build it all:
-<code bash>
-#!/usr/bin/env bash
-set -x
-module load compiler/gnu-4.4.7
-module load gsl/1.15
-module load fftw/3.3-avx
-module load fftw/3.3-avx-float
-module load libframe/8.04-2
-module load libmetaio/8.4
-module load python/2.6.6-ligo
-module load swig/2.0.11
-set -e
-# Make space for the test
-mkdir -p lalinference_test
-cd lalinference_test
-WORKDIR=${PWD}
-INSTALLDIR=${PWD}/opt
-# Download the lalsuite repository
-if [ -e lalsuite ]
-then
-  echo "Found a previous copy of lalsuite, updating it"
-  cd lalsuite
-  git pull
-  cd ..
-else
-  echo "Downloading a new copy of lalsuite"
-  git clone git://ligo-vcs.phys.uwm.edu/lalsuite.git
-fi
-# Condor-compile the lalsuite repository
-echo "Compiling lalsuite"
-cd lalsuite
-./00boot && ./configure --prefix=${INSTALLDIR} --enable-swig-python --disable-laldetchar && make -j8 install
-echo "Compiling GLUE and PYLAL"
-for F in ${WORKDIR}/opt/etc/*env.sh; do source $F; done
-cd pylal && python setup.py install --prefix=${INSTALLDIR}
-cd ../glue
-python setup.py install --prefix=${INSTALLDIR}
-cd ${WORKDIR}
-echo "Finished installing PYLAL and GLUE"
-</code>
 ===== LALInference Nest =====
@@ Line 425: / Line 376: @@
 Using ''--Nmcmc 100'':
 <code>
+Turbo Boost On: 3500 MHz
 real    15m17.527s
 user    15m14.425s
 sys     0m0.404s
+Turbo Boost Off: 2700 MHz
+real    19m38.241s
+user    19m35.629s
+sys     0m0.540s
 </code>
 and for ''--Nmcmc 200''
 <code>
+Turbo Boost On
 real    35m40.957s
 user    20m12.832s
@@ Line 443: / Line 401: @@
 sys     0m0.586s
 </code>
+Haswell node:
+<code>
+Turbo Boost Off
+real    20m52.627s
+user    20m52.242s
+sys     0m0.435s
+Turbo Boost Off: Icc -O3 -xHost
+real    12m43.876s
+user    12m43.007s
+sys     0m0.363s
+and for 24: 13m09s - 13m53s
+</code>
 ===== Heterodyne Pulsar =====
@@ Line 448: / Line 423: @@
 <code bash>
 cp -R /home/spxph/AcceptanceTestsHeterodyne/* .
-. /etc/lalapps-user-env.sh
+. $INSTALLDIR/etc/lalapps-user-env.sh
 time lalapps_heterodyne_pulsar --heterodyne-flag 0 --ifo H1 --pulsar J0000+0000 --param-file J0000+0000.par \
                                --sample-rate 16384 --resample-rate 1 --filter-knee 0.25 \
@@ Line 457: / Line 432: @@
 Running on Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz (Ivy Bridge) for the times ''931052708-931453056'' gives times of:
 <code>
+Turbo Boost On: 3500 MHz
 real    12m43.314s
 user    12m0.647s
 sys     0m34.930s
+Turbo Boost Off: 2700 MHz
+real 16m9.190s
+user 15m26.190s
+sys 0m35.840s
 </code>
 Raven Sandy Bridge node for the times: ''931052708-931453056'':
@@ Line 466: / Line 447: @@
 user    16m28.514s
 sys     1m26.912s
+</code>
+Raven Haswell node.
+<code>
+real    16m12.350s
+user    13m54.145s
+sys     1m12.344s
 </code>