zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Plots runtimes by analysis name and exechosts 

Useful to understand the performance of analysis jobs by hosts,
and to debug problematic nodes
This commit is contained in:
Mark DePristo 2011-12-07 09:24:47 -05:00
parent 5d2212bc8e
commit e17a1923fb
1 changed files with 31 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
public/R/scripts/org/broadinstitute/sting/queue/util/queueJobReport.R
View File

@ -12,7 +12,7 @@ if ( onCMDLine ) {
inputFileName = args[1] inputFileName = args[1]
outputPDF = args[2] outputPDF = args[2]
} else { } else {
inputFileName = "~/Desktop/broadLocal/GATK/unstable/wgs.jobreport.txt" inputFileName = "Q-8271@gsa2.jobreport.txt"
#inputFileName = "/humgen/gsa-hpprojects/dev/depristo/oneOffProjects/Q-25718@node1149.jobreport.txt" #inputFileName = "/humgen/gsa-hpprojects/dev/depristo/oneOffProjects/Q-25718@node1149.jobreport.txt"
#inputFileName = "/humgen/gsa-hpprojects/dev/depristo/oneOffProjects/rodPerformanceGoals/history/report.082711.txt" #inputFileName = "/humgen/gsa-hpprojects/dev/depristo/oneOffProjects/rodPerformanceGoals/history/report.082711.txt"
outputPDF = NA outputPDF = NA
@ -149,6 +149,35 @@ convertUnits <- function(gatkReportData) {
lapply(gatkReportData, convertGroup) lapply(gatkReportData, convertGroup)
} }
#
# Plots runtimes by analysis name and exechosts
#
# Useful to understand the performance of analysis jobs by hosts,
# and to debug problematic nodes
#
plotTimeByHost <- function(gatkReportData) {
fields = c("analysisName", "exechosts", "runtime")
runtimes = data.frame()
for ( report in gatkReportData ) {
runtimes = rbind(runtimes, report[,fields])
}
plotMe <- function(name, vis) {
p = ggplot(data=runtimes, aes(x=exechosts, y=runtime, group=exechosts, color=exechosts))
p = p + facet_grid(analysisName ~ .)
p = p + vis()
p = p + xlab("Job execution host")
p = p + opts(title = paste(name, "of job runtimes by analysis name and execution host"))
p = p + ylab(paste("Distribution of runtimes", RUNTIME_UNITS))
p = p + opts(axis.text.x=theme_text(angle=45, hjust=1, vjust=1))
print(p)
}
plotMe("Boxplot", geom_boxplot)
plotMe("Jittered points", geom_jitter)
}
# read the table # read the table
gatkReportData <- gsa.read.gatkreport(inputFileName) gatkReportData <- gsa.read.gatkreport(inputFileName)
@ -162,6 +191,7 @@ if ( ! is.na(outputPDF) ) {
plotJobsGantt(gatkReportData, T, F) plotJobsGantt(gatkReportData, T, F)
plotJobsGantt(gatkReportData, F, F) plotJobsGantt(gatkReportData, F, F)
plotProgressByTime(gatkReportData) plotProgressByTime(gatkReportData)
plotTimeByHost(gatkReportData)
for ( group in gatkReportData ) { for ( group in gatkReportData ) {
plotGroup(group) plotGroup(group)
} }