hi ulul thank you for explanation. I have googled the feature, and hortonworks said
This feature is a technical preview and considered under development. Do not use this feature in your production systems. can we use it in production env? 2015-02-15 20:15 GMT+08:00 Ulul <had...@ulul.org>: > Hi > > Actually it depends : in MR1 each mapper or reducer will be exezcuted in > its own JVM, in MR2 you can activate uberjobs that will let the framework > serialize small jobs' mappers and reducers in the applicationmaster JVM. > > Look for mapreduce.job.ubertask.* properties > > Ulul > > Le 15/02/2015 11:11, bit1...@163.com a écrit : > > Hi, Hadoopers, > > I am pretty newbie to Hadoop, I got a question: when a job runs, Will > each mapper or reducer task take up a JVM process or only a thread? > I hear that the answer is the Process. That is, say, one job contains 5 > mappers and 2 reducers , then there will be 7 JVM processes? > Thanks. > > ------------------------------ > bit1...@163.com > > >