[
https://issues.apache.org/jira/browse/MAHOUT-922?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13168754#comment-13168754
]
Hudson commented on MAHOUT-922:
-------------------------------
Integrated in Mahout-Quality #1251 (See
[https://builds.apache.org/job/Mahout-Quality/1251/])
MAHOUT-922:optional p, AB' tweaks, faster tests, style overhaul.
dlyubimov :
http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1213842
Files :
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/ABtDenseOutJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/BBtJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/BtJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/DenseBlockWritable.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/QJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDCli.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDPrototype.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDSolver.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/SparseRowBlockWritable.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/UJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/UpperTriangular.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/VJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/YtYJob.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/qr/GivensThinSolver.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/qr/GramSchmidt.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/qr/GrammSchmidt.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/qr/QRFirstStep.java
*
/mahout/trunk/core/src/main/java/org/apache/mahout/math/hadoop/stochasticsvd/qr/QRLastStep.java
*
/mahout/trunk/core/src/test/java/org/apache/mahout/math/hadoop/stochasticsvd/LocalSSVDSolverDenseTest.java
*
/mahout/trunk/core/src/test/java/org/apache/mahout/math/hadoop/stochasticsvd/LocalSSVDSolverSparseSequentialTest.java
*
/mahout/trunk/core/src/test/java/org/apache/mahout/math/hadoop/stochasticsvd/SSVDTestsHelper.java
> SSVD: ABt Job tweaks for extra sparse inputs
> --------------------------------------------
>
> Key: MAHOUT-922
> URL: https://issues.apache.org/jira/browse/MAHOUT-922
> Project: Mahout
> Issue Type: Improvement
> Components: Math
> Affects Versions: 0.6
> Reporter: Dmitriy Lyubimov
> Assignee: Dmitriy Lyubimov
> Fix For: 0.6
>
> Attachments: MAHOUT-922.patch, MAHOUT-922.patch, MAHOUT-922.patch
>
>
> Per tests on Sebastian's extremely sparse large inputs (4.5m x 4.5 m).
> AB' performance is still a bottleneck if one uses power iterations. For
> sufficiently sparse inputs it may turn out that mappers cannot form the
> entire blocked product in memory for Y_i. the Y_i block is going to be of
> size s x (k+p) where s is number of A rows read in a given split. in cases
> when A is extra sparse, such blocks may actually take more space than the A
> input. When this happens, s is constrained by -oh parameter and combiners and
> reducers get flooded by partial oh x (k+p) outer products and seem to have
> hard time to sort and shuffle them (especially high pressure on combiners has
> been seen).
> So, several improvements in this patch:
> -- present Y_i blocks as dense (they are beleived to be dense anyway, so
> keeping them as sparse just eats up RAM by sparse encoding, so at least twice
> as high blocks can actually be formed);
> -- eliminate combining completely. instead of persisting and sorting and
> summing up partial product in combiner, sum up map-side. if block height is
> still insufficient and cannot be extended due RAM constraints (unlikely for
> Sebastien's 4.5 x 4.5 mln case) just perform additional passes over B'. Since
> computation is cpu bound, additional passes over B' should not register.
> However, elimination of combiner phase for high load cases is probably going
> to have a dramatic effect.
> -- set max block height for Q'A and AB' separately instead of single -oh
> option. Their scaling seems to be quite different in terms of OOM danger. in
> my experiments Q'A blocking enters red zone at ~150,000 already whereas AB'
> block height can freely roam over a million easily for the same RAM. I
> provide 200,000 (~160Mb for k+p=100) as a default for AB' blocks which should
> be enough for Sebastien's 4.5 x 4.5 mln sparse case without causing more than
> one block.
> Miscellanea:
> -- Test run time: removed redundant tests and checks for SSVD. reduced test
> input size.
> -- Per Nathan's suggestion, p parameter is now optional, default is 15
> (computation time is cubic to it, so I want to be careful not to run it too
> high by default).
> Current patch branch work is here:
> https://github.com/dlyubimov/mahout-commits/tree/MAHOUT-922
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
