[ https://issues.apache.org/jira/browse/GIRAPH-873?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14015086#comment-14015086 ]
Hudson commented on GIRAPH-873: ------------------------------- ABORTED: Integrated in Giraph-trunk-Commit #1442 (See [https://builds.apache.org/job/Giraph-trunk-Commit/1442/]) GIRAPH-873 : Specialized edge stores (pavanka: http://git-wip-us.apache.org/repos/asf?p=giraph.git&a=commit&h=9cedc7d76f2bbe52b3d1cc4caf8024e730266f83) * CHANGELOG * giraph-core/src/main/java/org/apache/giraph/edge/SimpleEdgeStore.java * giraph-core/src/main/java/org/apache/giraph/edge/EdgeStore.java * giraph-core/src/main/java/org/apache/giraph/edge/primitives/package-info.java * giraph-core/src/main/java/org/apache/giraph/edge/InMemoryEdgeStoreFactory.java * giraph-core/src/main/java/org/apache/giraph/edge/AbstractEdgeStore.java * giraph-core/src/main/java/org/apache/giraph/edge/primitives/IntEdgeStore.java * giraph-core/src/main/java/org/apache/giraph/edge/EdgeStoreFactory.java * giraph-core/src/main/java/org/apache/giraph/conf/GiraphConstants.java * giraph-core/src/main/java/org/apache/giraph/edge/primitives/LongEdgeStore.java * giraph-core/src/main/java/org/apache/giraph/comm/ServerData.java * giraph-core/src/main/java/org/apache/giraph/conf/ImmutableClassesGiraphConfiguration.java > Specialized edge stores > ----------------------- > > Key: GIRAPH-873 > URL: https://issues.apache.org/jira/browse/GIRAPH-873 > Project: Giraph > Issue Type: Improvement > Affects Versions: 1.1.0 > Reporter: Craig Muchinsky > Assignee: Pavan Kumar > Fix For: 1.1.0 > > Attachments: GIRAPH-873-2.patch, GIRAPH-873.patch, > GIRAPH-873_refactor.patch > > > While doing some performance tuning I discovered that loading the edge store > can be a very expensive operation. Similar to GIRAPH-704, the use of > primitive maps can provide significant performance benefit. Part of the > benefit comes with the lower memory overhead associated with the primitive > maps however the larger benefit comes with the fact that you don't have to > release and reconstruct the vertexId object every time a new vertex is > encountered. > When processing a large graph with 4B vertices and 5B edges (3B of the edges > loaded via EdgeInputFormat) the worker edge requests were taking ~15 seconds > each, but after implementing the above suggestions that number dropped down > sub-second. -- This message was sent by Atlassian JIRA (v6.2#6252)