These scripts are intended to enable the retrieval of information from map, shuffle and reduce tasks, regarding execution time.
- Linux
- python
- gnuplot
python generate_graphics.py [1] [2] [3] [4] [5]
[1] job tracker file path [2] task tracker file path [3] job id prefix. Ex.: in job_201101221816_0005 the prefix is 201101221816 [4] first job id [5] last job id (this enables the processing of a range of jobs)
./generate_graphics.py hadoop-root-jobtracker-graphite-2.nancy.grid5000.fr.log hadoop-root-tasktracker-graphite-2.nancy.grid5000.fr.log 201501100904 1 30