| clear query|facets|time |
Search criteria: .
Results from 11 to 17 from
17 (0.195s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: Performance of direct vs indirect shuffling - MapReduce - [mail # user]
|
|
... Have someone ever considered a general data transfer service bundled with YARN? So other applications(rather than MR) can also benefits from it. The data transfer service looks like re...
|
|
|
Author: Binglin Chang,
2011-12-21, 06:07
|
|
|
[MAPREDUCE-3246] Make Task extensible to support modifications of Task or even alternate programming paradigms - MapReduce - [issue]
|
|
...One of MRv2's goal is to support alternate programming paradigms, but building a application using YARN from the bottom is not trivial. In fact most component of MapReduce can be reused, mos...
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-3246
Author: Binglin Chang,
2011-11-27, 14:37
|
|
|
[MAPREDUCE-2541] Race Condition in IndexCache(readIndexFileToCache,removeMap) causes value of totalMemoryUsed corrupt, which may cause TaskTracker continue throw Exception - MapReduce - [issue]
|
|
...The race condition goes like this:Thread1: readIndexFileToCache() totalMemoryUsed.addAndGet(newInd.getSize())Thread2: removeMap() totalMemoryUsed.addAndGet(-info.getSize());When SpillR...
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-2541
Author: Binglin Chang,
2011-11-15, 00:48
|
|
|
[MAPREDUCE-3383] Duplicate job.getOutputValueGroupingComparator() in ReduceTask - MapReduce - [issue]
|
|
...This is probably just a small error by mistake....
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-3383
Author: Binglin Chang,
2011-11-09, 08:11
|
|
|
[MAPREDUCE-2910] Allow empty MapOutputFile segments - MapReduce - [issue]
|
|
...As the scale of cluster and job get larger, we see a lot of empty partitions in MapOutputFile due to large reduce numbers or partition skew. When map output compression is enabled, empty map...
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-2910
Author: Binglin Chang,
2011-10-31, 04:57
|
|
|
[MAPREDUCE-3086] Supporting range scan using TFile, TotalOrderPartitioner and partition index - MapReduce - [issue]
|
|
...Hive/HBase already has similar or more powerful functionality, but using hive/hbase is overkill or inconvenient for some cases, so add some lightweight utility classes to only support range ...
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-3086
Author: Binglin Chang,
2011-10-12, 09:01
|
|
|
[MAPREDUCE-2872] Optimize TaskTracker memory usage - MapReduce - [issue]
|
|
...We observe high memory usage of framework level components on slave node, mainly TaskTracker & Child, especially for large clusters. To be clear at first, large jobs with 10000-100000 ma...
|
|
|
http://issues.apache.org/jira/browse/MAPREDUCE-2872
Author: Binglin Chang,
2011-08-30, 08:46
|
|
|
|
