Re:Re: how to estimate job run time?

Thanks, Moustafa Gabe. It's really a good project,I downloaded it,I will try it sometime. thank you. Best Regards, Junqing Zhou At 2011-06-14 21:29:02，"Moustafa Gaber" wrote: Actually, It is recently a research point. However, Starfish is a good start in this area: http://www.cs.duke.edu/star

Re: mapred.child.java.opts question

No it does not. Sent from my iPhone On Jun 14, 2011, at 11:22 AM, Jeff Bean wrote: > Does your class use GenericOptionsParser (does it implement Tool, and does it > call ToolRunner.run(), for example? > > Jeff > > On Tue, Jun 14, 2011 at 8:47 AM, Alex Kozlov wrote: > There might be differen

Re: mapred.child.java.opts question

Does your class use GenericOptionsParser (does it implement Tool, and does it call ToolRunner.run(), for example? Jeff On Tue, Jun 14, 2011 at 8:47 AM, Alex Kozlov wrote: > There might be different reasons why this parameter is not passed to the > slave JVM: for example, it might have been decl

unsubscribe

unsubscribe

Re: mapred.child.java.opts question

There might be different reasons why this parameter is not passed to the slave JVM: for example, it might have been declared final. Do you see the correct parameter in your job xml file (to be found in the JT UI or in the slave local FS)? You can also see the passed parameters if you do `ps aux`

Re: mapred.child.java.opts question

Sorry about the last message. Here we go again: I am trying to pass this option with my job as: hadoop jar -Dmapred.child.java.opts=-Xmx1000m -conf But I still get the error: "Error: Java Heap Space" for all the task trackers. Could somebody advice how can I make this value propagate to al

mapred.child.java.opts question

Hi, I am trying to pass this option with my

Re: can't compile the mapreduce project in eclipse

Did you add all dependencies of the source code? El 6/14/2011 10:32 AM, Erix Yao escribió: hi I checked out the source code from http://svn.apache.org/repos/asf/hadoop/mapreduce/tags/release-0.21.0 and execute ant compile eclipse-files , but after import the project into eclipse, I found

can't compile the mapreduce project in eclipse

hi I checked out the source code from http://svn.apache.org/repos/asf/hadoop/mapreduce/tags/release-0.21.0 and execute ant compile eclipse-files , but after import the project into eclipse, I found the error as below: Description Resource Path Location Type Project 'mapreduce' is missing requir

Re: Delimiter selection for Sequence Files

Thanks Joe fit the reply ! "@@##@@" looks like a big value for a delimiter. I will also choose something like a hex number so that it does not appear in the data. Sent from my iPhone On Jun 13, 2011, at 5:33 PM, Joe Stein wrote: > I have had quite a few data sets that I have had no idea if my

Re: how to estimate job run time?

Actually, It is recently a research point. However, Starfish is a good start in this area: http://www.cs.duke.edu/starfish/ 2011/6/14 周俊清 <2ho...@163.com> > hello everyone, > i want to estimate the hadoop job's run time or the single map/reduce run > time for my job scheduling algorithm ? It's a

how to estimate job run time?

hello everyone, i want to estimate the hadoop job's run time or the single map/reduce run time for my job scheduling algorithm ? It's anyone doing about this work? thank you for concerning,any advice will be appreciated. Junqing Zhou 2ho...@163.com