]
Galder Zamarreño updated ISPN-49:
---------------------------------
Attachment: run4.zip
Here's some graphs after implementing a solution based purely on ObjectTable and
pooling marshaller with thread locals. Performance overall looks much better but I'm
intrigued about 4-6 node async repl perf with JBMAR. This is currently being
investigated.
Perf test using JBoss Marshalling marshaller layer
--------------------------------------------------
Key: ISPN-49
URL:
https://jira.jboss.org/jira/browse/ISPN-49
Project: Infinispan
Issue Type: Task
Components: RPC
Reporter: Galder Zamarreño
Assignee: Galder Zamarreño
Fix For: 4.0.0.BETA1, 4.0.0.GA
Attachments: round-lab-0.zip, round-lab-1.zip, run1.zip, run2.zip, run3.zip,
run4.zip
Run performance comparisons between Infinispan using home grown marshalling framework
and the one based on JBoss Marshalling.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: