Adding gnuplot and data files for disaggregation blog post

apache · Mar 5, 2019 · 3c0c96bc7df96158f8a27168976b0ff354cef4b9 · 3c0c96b
1 parent c9ab761
commit 3c0c96bc7df96158f8a27168976b0ff354cef4b9
Showing 36 changed files with 50,715 additions and 0 deletions.
diff --git a/site/_plots/blog/disaggregation/data-distribution/cdf-plot.pdf b/site/_plots/blog/disaggregation/data-distribution/cdf-plot.pdf
diff --git a/site/_plots/blog/disaggregation/data-distribution/cdf-plot.svg b/site/_plots/blog/disaggregation/data-distribution/cdf-plot.svg
diff --git a/site/_plots/blog/disaggregation/data-distribution/clean-sort-generate-cdf.sh b/site/_plots/blog/disaggregation/data-distribution/clean-sort-generate-cdf.sh
@@ -0,0 +1,22 @@
+# which column 
+if [ $# -ne 2 ]; then 
+	echo "tell me file name and col number"
+	exit 1
+fi
+k=$2
+file=$1 
+
+# sort it 
+sort --key $k -n $file > $file-sort 
+
+# now we generate CDF 
+totalLines=$(cat $file-sort | wc -l)
+echo "total line numbers: $totalLines" 
+rm ./xx
+for i in `seq 1 100`;
+do 
+   lx=$(( $i * $totalLines ))
+   lxx=$(( $lx / 100 )) # you get the line number here 
+   sed -n "${lxx}p" $file-sort >> ./xx
+done    
+cat ./xx | awk '{print $1,"\t",NR}' > cdf-$file