directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Emmanuel Lecharny (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DIRSERVER-959) We nedd a global cache
Date Tue, 26 Feb 2013 10:10:13 GMT

    [ https://issues.apache.org/jira/browse/DIRSERVER-959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13586985#comment-13586985
] 

Emmanuel Lecharny commented on DIRSERVER-959:
---------------------------------------------

We have added a CacheService class for that purpose.
                
> We nedd a global cache 
> -----------------------
>
>                 Key: DIRSERVER-959
>                 URL: https://issues.apache.org/jira/browse/DIRSERVER-959
>             Project: Directory ApacheDS
>          Issue Type: Improvement
>            Reporter: Emmanuel Lecharny
>             Fix For: 2.1.0
>
>
> Currently, we only have local caches assigned to each partition :
> - a so called "global" cache which stores the partition entries
> - a cache for each declared index
> We also have real global caches :
> - a cache for referrals
> - a cache for ACIs (tuples)
> - a cache for subentries
> - a cache for principals
> - a cache for Groups 
> - a cache for notAlias entries
> Up to this point, it becomes to be a full mess. We have so many different caches, with
so many different configurations, that it's almost impossible to know how to correctly tune
the server. Moreover, we may have empty caches when other are heavily hit.
> After having discussed about this situation with Alex, we think wit would be better to
have a global cache shared between all the partitions, but with added functionnalities :
> - the possibility to balance the cache between partitions (for instance, 20% for 'ou=system',
80% for 'dc=example, dc=com'
> - the possibility to change this balance dynamically, depending on the actual hit rate
> - the possibility to dedicate a part of this cach to pin some entries in memory (like
the schema partition entries, for instance)
> - the possibility to increase or decrease the cache size dynamically, without having
to stop and restart the server
> It would be also a good idea to be able to have a global Interface for all the cache
we have, and may be a cache manager instead of a specific implementation for each cache
> Feel free to add comments in this JIRA !

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message