[
https://issues.apache.org/jira/browse/SYSTEMML-1072?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15623765#comment-15623765
]
Glenn Weidner commented on SYSTEMML-1072:
-----------------------------------------
Thank you Matthias for the feedback and pointer to SYSTEMML-1026! I've
repeated the runMultiLogReg dense 8G experiment and got consistent results for
consecutive runs:
-- Running runMultiLogReg on 1M_1k_dense (all configs)
MultiLogReg train ict=0 on mbperftest/multinomial/X1M_1k_dense_k150: 141
MultiLogReg train ict=1 on mbperftest/multinomial/X1M_1k_dense_k150: 169
MultiLogReg train ict=2 on mbperftest/multinomial/X1M_1k_dense_k150: 414
-- Running runMultiLogReg on 1M_1k_dense (all configs)
MultiLogReg train ict=0 on mbperftest/multinomial/X1M_1k_dense_k150: 145
MultiLogReg train ict=1 on mbperftest/multinomial/X1M_1k_dense_k150: 165
MultiLogReg train ict=2 on mbperftest/multinomial/X1M_1k_dense_k150: 417
My sparkDML.sh is configured such that spark-submit uses below options:
$SPARK_HOME/bin/spark-submit
--master yarn-client
--driver-memory 20G
--num-executors 6
--executor-memory 55G
--executor-cores 24
--driver-java-options "-server -Xms20g"
--conf spark.executor.extraJavaOptions="-server -Xmn5500m"
I will repeat binomial runMultiLogReg for 80G. If consistent results, then I
think I can close this JIRA. I had opened since there seemed to be a bigger
difference when compared to Oct 12/13 times which might not be appropriate
comparison.
> Perftest: MultiLogReg performance variation
> -------------------------------------------
>
> Key: SYSTEMML-1072
> URL: https://issues.apache.org/jira/browse/SYSTEMML-1072
> Project: SystemML
> Issue Type: Bug
> Reporter: Glenn Weidner
>
> Times (8G, k150):
> SPARK RUN MULTINOMIAL EXPERIMENTS: Thu Oct 27
> -- Running runMultiLogReg on 1M_1k_dense (all configs)
> MultiLogReg train ict=0 on mbperftest/multinomial/X1M_1k_dense_k150: 141
> MultiLogReg train ict=1 on mbperftest/multinomial/X1M_1k_dense_k150: 166
> MultiLogReg train ict=2 on mbperftest/multinomial/X1M_1k_dense_k150: 417
> SPARK RUN MULTINOMIAL EXPERIMENTS: Wed Oct 12
> -- Running runMultiLogReg on 1M_1k_dense (all configs)
> MultiLogReg train ict=0 on mbperftest/multinomial/X1M_1k_dense_k150: 127
> MultiLogReg train ict=1 on mbperftest/multinomial/X1M_1k_dense_k150: 151
> MultiLogReg train ict=2 on mbperftest/multinomial/X1M_1k_dense_k150: 354
> Times (80G, dense):
> SPARK RUN BINOMIAL EXPERIMENTS: Wed Oct 26
> -- Running runMultiLogReg on 10M_1k_dense (all configs):
> MultiLogReg train ict=0 on mbperftest/binomial/X10M_1k_dense: 143
> MultiLogReg train ict=1 on mbperftest/binomial/X10M_1k_dense: 147
> MultiLogReg train ict=2 on mbperftest/binomial/X10M_1k_dense: 138
> SPARK RUN BINOMIAL EXPERIMENTS: Wed Oct 12
> -- Running runMultiLogReg on 10M_1k_dense (all configs):
> MultiLogReg train ict=0 on mbperftest/binomial/X10M_1k_dense: 90
> MultiLogReg train ict=1 on mbperftest/binomial/X10M_1k_dense: 84
> MultiLogReg train ict=2 on mbperftest/binomial/X10M_1k_dense: 94
> Times (80G, k150):
> SPARK RUN MULTINOMIAL EXPERIMENTS: Wed Oct 26
> -- Running runMultiLogReg on 10M_1k_dense (all configs)
> MultiLogReg train ict=0 on mbperftest/multinomial/X10M_1k_dense_k150: 1300
> MultiLogReg train ict=1 on mbperftest/multinomial/X10M_1k_dense_k150: 1265
> MultiLogReg train ict=2 on mbperftest/multinomial/X10M_1k_dense_k150: 1677
> SPARK RUN MULTINOMIAL EXPERIMENTS: Wed Oct 12
> -- Running runMultiLogReg on 10M_1k_dense (all configs)
> MultiLogReg train ict=0 on mbperftest/multinomial/X10M_1k_dense_k150: 1639
> MultiLogReg train ict=1 on mbperftest/multinomial/X10M_1k_dense_k150: 1790
> MultiLogReg train ict=2 on mbperftest/multinomial/X10M_1k_dense_k150: 2541
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
