plot historical timings over time, closes h2oai#46

history.Rmd

@@ -0,0 +1,42 @@
+---
+title: "History of db-benchmark"
+output:
+  html_document:
+    self_contained: no
+    toc: false
+---
+
+```{r init, echo=FALSE}
+knitr::opts_chunk$set(echo=FALSE, cache=FALSE)
+library(lattice)
+source("report.R")
+report_status_file = get_report_status_file()
+ld = time_logs()
+```
+
+## Timings over time
+
+### groupby
+
+0% NAs, unsorted
+
+```{r hist_groupby_plot, fig.width=8, fig.height=48}
+stopifnot(length(unique(ld$nodename))==1L)
+p = sapply(setNames(nm=as.character(unique(ld$solution))), simplify = FALSE, function(s)
+  lattice::xyplot(time_sec_1 ~ ibatch | k+in_rows, ld[nasorted=="0% NAs, unsorted"],
+                type="l", grid=TRUE, groups=question,
+                subset=solution==s, main=s,
+                xlab = "benchmark run",
+                ylab = "seconds",
+                scales=list(y=list(
+                  relation="free",
+                  limits=rep(ld[solution==s, .(ylim=max(c(0, time_sec_1), na.rm=TRUE)), in_rows][ylim>0, list(list(c(0, ylim))), in_rows]$V1, each=3)
+                )),
+                auto.key=list(points=FALSE, lines=TRUE))
+)
+sapply(seq_along(p), function(i) print(p[[i]], split=c(1, i, 1, length(p)), more=i!=length(p))) -> nul
+```
+
+```{r set_success_state, include=FALSE}
+#cat("history\n", file=report_status_file, append=TRUE)
+```

run.sh

@@ -33,17 +33,18 @@ if [[ "$DO_UPGRADE" == true && "$RUN_SOLUTIONS" =~ "spark" ]]; then ./spark/init
 Rscript ./launcher.R
 
 # publish report for all tasks
-rm -f rmarkdown-index.out rmarkdown-tech.out
+rm -f rmarkdown-index.out rmarkdown-tech.out rmarkdown-history.out
 rm -rf public
 rm -f report-done
 Rscript -e 'rmarkdown::render("index.Rmd", output_dir="public")' > ./rmarkdown-index.out 2>&1 && echo "# Benchmark report produced"
+Rscript -e 'rmarkdown::render("history.Rmd", output_dir="public")' > ./rmarkdown-history.out 2>&1 && echo "# Benchmark history report produced"
 Rscript -e 'rmarkdown::render("tech.Rmd", output_dir="public")' > ./rmarkdown-tech.out 2>&1 && echo "# Benchmark tech report produced"
 
 # publish benchmark, only if reports successfully generated (groupby, tech), token file exists
 rm -rf db-benchmark.gh-pages
 $DO_PUBLISH \
   && [ -f ./report-done ] \
-  && [ $(wc -l report-done | awk '{print $1}') -eq 2 ] \
+  && [ $(wc -l report-done | awk '{print $1}') -eq 3 ] \
   && [ -f ./token ] \
   && ((./publish.sh && echo "# Benchmark results has been published") || echo "# Benchmark publish script failed")
 

tech.Rmd

@@ -2,7 +2,7 @@
 title: "Technical measures of db-benchmark"
 output:
   html_document:
-    self_contained: yes
+    self_contained: no
     toc: true
 ---