Yes, I'm still using version 0.5, the plan is to verify it can work on 0.5 and
get some benchmark first then moving forward to 0.6, Sebastian, do you think
it's a problem related to Mahout? (Not Hadoop?) And do you think 0.6 will bring
us a huge performance increase? Thanks. CheersRamon
> Date: Wed, 19 Oct 2011 10:20:24 +0200
> From: [email protected]
> To: [email protected]
> Subject: Re: Exception during running
> RowSimilarityJob-Mapper-EntriesToVectorsReducer job
>
> It seems like you're still not using Mahout 0.6. Please use the latest
> version and apply appropriate down sampling to your input data. You
> should also try to get access to a cluster with more than 2 machines.
>
> --sebastian
>
> On 19.10.2011 10:16, WangRamon wrote:
> >
> >
> >
> >
> > Hi Guys I'm continuing running the test case with a 1GB data file which
> > contains 600000 users and 2000000 items, all the jobs are running in a 2
> > two nodes cluster, each node has 32GB RAM and 8 core CPU, the
> > RecommenderJob running until it reach
> > RowSimilarityJob-Mapper-EntriesToVectorsReducer Job, see below for the
> > error log: 11/10/18 23:09:34 INFO mapred.JobClient: map 11% reduce 1%
> > 11/10/18 23:12:46 INFO mapred.JobClient: map 11% reduce 2%
> > 11/10/18 23:13:55 INFO mapred.JobClient: map 12% reduce 2%
> > 11/10/18 23:18:22 INFO mapred.JobClient: map 13% reduce 2%
> > 11/10/18 23:22:50 INFO mapred.JobClient: map 14% reduce 2%
> > 11/10/18 23:27:08 INFO mapred.JobClient: map 15% reduce 2%
> > 11/10/18 23:28:15 INFO mapred.JobClient: map 15% reduce 3%
> > 11/10/18 23:31:42 INFO mapred.JobClient: map 16% reduce 3%
> > 11/10/18 23:33:36 INFO mapred.JobClient: Task Id :
> > attempt_201110181002_0007_r_000000_0, Status : FAILED
> > java.io.IOException: Task: attempt_201110181002_0007_r_000000_0 - The
> > reduce copier failed
> > at org.apache.hadoop.mapred.ReduceTask.run(ReduceTask.java:380)
> > at org.apache.hadoop.mapred.Child.main(Child.java:170)
> > Caused by: java.io.IOException: Intermediate merge failed
> > at
> > org.apache.hadoop.mapred.ReduceTask$ReduceCopier$InMemFSMergeThread.doInMemMerge(ReduceTask.java:2576)
> > at
> > org.apache.hadoop.mapred.ReduceTask$ReduceCopier$InMemFSMergeThread.run(ReduceTask.java:2501)
> > Caused by: java.lang.RuntimeException: java.io.EOFException
> > at
> > org.apache.hadoop.io.WritableComparator.compare(WritableComparator.java:103)
> > at org.apache.hadoop.mapred.Merger$MergeQueue.lessThan(Merger.java:373)
> > at org.apache.hadoop.util.PriorityQueue.upHeap(PriorityQueue.java:123)
> > at org.apache.hadoop.util.PriorityQueue.put(PriorityQueue.java:50)
> > at org.apache.hadoop.mapred.Merger$MergeQueue.merge(Merger.java:447)
> > at org.apache.hadoop.mapred.Merger$MergeQueue.merge(Merger.java:381)
> > at org.apache.hadoop.mapred.Merger.merge(Merger.java:107)
> > at org.apache.hadoop.mapred.Merger.merge(Merger.java:93)
> > at
> > org.apache.hadoop.mapred.ReduceTask$ReduceCopier$InMemFSMergeThread.doInMemMerge(ReduceTask.java:2551)
> > ... 1 more
> > Caused by: java.io.EOFException
> > at java.io.DataInputStream.readByte(DataInputStream.java:250)
> > at org.apache.mahout.math.Varint.readUnsignedVarInt(Varint.java:159)
> > at org.apache.mahout.math.Varint.readSignedVarInt(Varint.java:140)
> > at
> > org.apache.mahout.math.hadoop.similarity.SimilarityMatrixEntryKey.readFields(SimilarityMatrixEntryKey.java:64)
> > at
> > org.apache.hadoop.io.WritableComparator.compare(WritableComparator.java:97)
> > ... 9 more11/10/18 23:33:37 INFO mapred.JobClient: map 16% reduce 0%
> > 11/10/18 23:35:57 INFO mapred.JobClient: map 17% reduce 0% I googled a lot
> > and find i should increase "mapred.reduce.tasks" property in Hadoop, so I
> > set it to 8 in my environment and restart this Job only, so far so good,
> > the job is still running by now, but it's still a little slow, so here
> > comes my questions: 1) does it be so slow for this job
> > RowSimilarityJob-Mapper-EntriesToVectorsReducer Job ?2) what does this
> > property "mapred.reduce.tasks" do? And why it can effect
> > RowSimilarityJob-Mapper-EntriesToVectorsReducer Job ? (Maybe i should ask
> > this 2nd question in hadoop user list... but i think people here are both
> > pro at hadoop :) )3) what can i do to increase the speed for this job? Any
> > ideas? Thanks in advance! Ramon
>
