Details
-
Improvement
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
-
None
-
None
Description
As discussed in GIRAPH-12(http://goo.gl/CE32U), I think that there is a potential problem to cause out of memory when the rate of message generation is higher than the rate of message flush (or network bandwidth).
To overcome this problem, we need more eager strategy for message flushing or some approach to spill messages into disk.
The below link is Dmitriy's suggestion.
https://issues.apache.org/jira/browse/GIRAPH-12?focusedCommentId=13116253&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13116253
Attachments
Attachments
Issue Links
- is related to
-
GIRAPH-185 Improve concurrency of putMsg / putMsgList
- Resolved
- relates to
-
GIRAPH-153 HBase/Accumulo Input and Output formats
- Resolved
-
GIRAPH-78 Be smarter about multiple instances of the same vertex
- Resolved
-
GIRAPH-249 Move part of the graph out-of-core when memory is low
- Resolved
-
GIRAPH-12 Investigate communication improvements
- Resolved