Bug 1152934 - TopologyAwareConsistentHashFactory is slow for large cluster
Summary: TopologyAwareConsistentHashFactory is slow for large cluster
Keywords:
Status: CLOSED EOL
Alias: None
Product: JBoss Data Grid 6
Classification: JBoss
Component: Infinispan
Version: 6.3.1
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: ---
Assignee: Tristan Tarrant
QA Contact: Martin Gencur
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-10-15 08:35 UTC by Takayoshi Kimura
Modified: 2024-05-01 00:21 UTC (History)
1 user (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2024-05-01 00:21:08 UTC
Type: Bug
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker ISPN-4841 0 Major Open TopologyAwareConsistentHashFactory is slow for large cluster 2020-06-18 23:27:15 UTC

Description Takayoshi Kimura 2014-10-15 08:35:14 UTC 
Observed 100% CPU usage for a long time on coordinator node when booting 500 nodes with 500 caches defined.

It looks like the TopologyAwareConsistentHashFactory performs O(n^2), it has double loop for all Machines. It takes 50 sec to compute rebalance with 1 cache 500 nodes. This calculation is performed on every cache, so it eats 25000 sec CPU times with 500 nodes 500 caches.

The hprof shows 90% of the time is consumed in the TopologyInfo.computeMaxSegmentsForMachine().
Comment 2 Dan Berindei 2014-10-15 09:48:57 UTC 
Takayoshi, have you seen the perf problems only with TopologyAwareConsistentHashFactory? Have you also tested with TopologyAwareSyncConsistentHashFactory?
Comment 3 Takayoshi Kimura 2014-10-15 13:23:58 UTC 
Yes it's TopologyAwareConsistentHashFactory only, it took 55 sec with 500 nodes. The Sync one took only 2 sec, not a problem so far.
Note You need to log in before you can comment on or make changes to this bug.