[ https://jira.jboss.org/jira/browse/ISPN-59?page=com.atlassian.jira.plugin... ] Galder Zamarreño updated ISPN-59: --------------------------------- Attachment: org.infinispan.marshall.MarshallersTest-v2-onlyobjecttable_and_version=0.xml There's even more squeezing that can be done here. If we set JBMAR version to 0, JBMAR does not add to version info to each object it actually writes. We need a version for each call to Marshaller.objectToObjectStream(), not for each individual object that a single call to objectToObjectStream marshalls. So, if we set version to 0, the number of tests failing due to JBMAR serializing less bytes than home grown drastically reduces. Only one test fails now: testSetMarshalling: JBoss Marshaller should write less bytes: bytesJBoss=64, bytesHome=58 The other tests failing are related to home grown marshalling issues (similar to ISPN-80) or the MarshalledValue issue in ISPN-79. Bottom line: JBMAR is looking better and now the payloads are almost always smaller. -- 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