Can be the overhead added by Hadoop that big? I mean I have over 30000 small tasks (about one minute), each one starting its own JVM.
Ok. I have set the number on maps to about 1760 (11 nodes * 16
cores/node * 10 as recommended by Hadoop documentation) and my job still
takes several hours to run instead of one.
- Re: Processing 10MB files in Hadoop Jeff Zhang
- Re: Processing 10MB files in Hadoop CubicDesign
- Re: Processing 10MB files in Hadoop Jeff Zhang
- Re: Processing 10MB files in Hadoop CubicDesign
- Re: Processing 10MB files in Ha... Jeff Zhang
- Re: Processing 10MB files i... Jason Venner
- Re: Processing 10MB files i... CubicDesign
- Re: Processing 10MB files i... Aaron Kimball
- Re: Processing 10MB files i... CubicDesign
- Re: Processing 10MB files i... Aaron Kimball
- Re: Processing 10MB files i... CubicDesign
- Re: Processing 10MB files i... Patrick Angeles
- Re: Processing 10MB files i... CubicDesign
- Re: Processing 10MB files in Hadoop CubicDesign
- Good idea to run NameNode and JobTracker on same mach... Raymond Jennings III
- Re: Good idea to run NameNode and JobTracker on ... Jeff Zhang
- Re: Good idea to run NameNode and JobTracker on ... Yongqiang He
- Hadoop 0.20 map/reduce Failing for old API Arv Mistry
- Re: Hadoop 0.20 map/reduce Failing for o... Rekha Joshi
- RE: Hadoop 0.20 map/reduce Failing f... Arv Mistry
- Re: Hadoop 0.20 map/reduce Fail... Edward Capriolo