[JBoss JIRA] Created: (EJBTHREE-2052) Use an L1 Cache for SFSB context stored in a DIST mode Infinispan
by Brian Stansberry (JIRA)
Use an L1 Cache for SFSB context stored in a DIST mode Infinispan
-----------------------------------------------------------------
Key: EJBTHREE-2052
URL: https://jira.jboss.org/jira/browse/EJBTHREE-2052
Project: EJB 3.0
Issue Type: Sub-task
Components: cache, Clustering
Reporter: Brian Stansberry
With Infinispan DIST, an SFSB context may not be stored in Infinispan on the local node. (Even if it's originally stored locally, it may move due to a rehash following a topology change).
Keep an L1 cache on the node that's servicing the session to improve read performance.
The session ownership system discussed in EJBTHREE-2051 can be used to detect failover and the need to invalidate the session out of L1. The locking system should return whether the call required obtaining the lock from a remote node; if yes that implies failover and an invalid L1.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
15 years, 9 months
[JBoss JIRA] Created: (EJBTHREE-2050) Ensure ids for newly created SFSB contexts will hash locally in Infinispan
by Brian Stansberry (JIRA)
Ensure ids for newly created SFSB contexts will hash locally in Infinispan
--------------------------------------------------------------------------
Key: EJBTHREE-2050
URL: https://jira.jboss.org/jira/browse/EJBTHREE-2050
Project: EJB 3.0
Issue Type: Sub-task
Components: cache, Clustering
Reporter: Brian Stansberry
Assignee: Paul Ferraro
If Infinispan DIST is used, at least one of the cache locations for an SFSB should be the node that created it. This way in the normal case where there are 2 copies, a single unicast to the backup node is sufficient to replicate. If the cache key (the SFSBContext id) hashes to 2 remote nodes, then every cache write will require 2 unicasts.
The problem here is unlike web sessions, the id for an SFSB context is not generated by the caching layer; it's generated by StatefulBeanContext itself. So the web session strategy of having a background thread occasionally populate a pool of ids that are known to map locally (using the consistent hash algorithm Mircea Marcus is going to expose for ISPN-232) won't work.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
15 years, 9 months