[ 
https://issues.apache.org/jira/browse/JCR-3064?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13122020#comment-13122020
 ] 

Alex Parvulescu commented on JCR-3064:
--------------------------------------

being in the neighborhood, I did some profiling on ConcurrentReadTest using 
visualvm (hopefully the formatting will be friendly enough)
I'vve added the number of test runs next to the JR version.

# ConcurrentReadTest                                min     10%     50%     90% 
    max
2.2(431)                                                      18      34        
  84     164    1398

# ConcurrentReadTest                                 min     10%     50%     
90%     max
2.3(102)                                                       63     305       
497     645    1450

# ConcurrentReadTest(w parentId optim)       min     10%     50%     90%     max
2.3(130)                                                       12     142       
380     624    1421


Top 5 invocations (with the help of visualvm):
(method / self time / invocations)

JR 2.3
o.a.j.core.security.authorization.acl.CachingEntryCollector.getEntries()        
      656816 ms (14.3%) 3223
o.a.j.core.security.authorization.acl.CachingEntryCollector.getParentId()       
      510579 ms (9.3%)  3104
o.a.j.core.security.authorization.acl.EntryCollector.collectEntries()           
          1047 ms (0.3%)         902
o.a.j.core.security.authorization.acl.CompiledPermissionsImpl.canRead()         
  1009 ms (0.3%)         804
o.a.j.core.ItemManager.canRead()                                                
                       0.000 ms (0%)      803

JR 2.2:
o.a.j.core.state.DefaultISMLocking.acquireReadLock()                            
        299244 ms (14.4%)       2184
o.a.j.core.SessionImpl.getURI()                                                 
                104227 ms (5%)            833
o.a.j.spi.commons.name.RelativePath.getElements()                               
          79294 ms (3.8%)       745
o.a.j.core.cache.ConcurrentCache.get()                                          
                  53731 ms (2.6%)       677
o.a.j.spi.commons.name.PathMap$Element.getPathElement()                   51599 
ms (2.5%)       475


I'll attach some screenshots of the visualvm output, for some nice visuals.
                
> Concurrent access performance drop
> ----------------------------------
>
>                 Key: JCR-3064
>                 URL: https://issues.apache.org/jira/browse/JCR-3064
>             Project: Jackrabbit Content Repository
>          Issue Type: Bug
>            Reporter: Jukka Zitting
>            Priority: Blocker
>             Fix For: 2.4
>
>         Attachments: ConcurrentReadTest.png, ConcurrentReadWriteTest.png
>
>
> Our performance tests show a pretty bad drop in concurrent access performance 
> (both read and write) in the latest trunk when compared to Jackrabbit 2.2. We 
> need to track down the cause and fix it before the 2.3 release.

--
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

        

Reply via email to