On 19/07/2019 02:22, Yi Yang (杨燚)-云服务集团 wrote: > Robert, the data in https://pantheon.tech/opendaylight-neon-sr1-sr2/ looks > amazing, can you show us those gerrit changes? How can we double check those > data you got?
For Neon that's ... https://git.opendaylight.org/gerrit/#/c/82315/ https://git.opendaylight.org/gerrit/#/c/82316/ https://git.opendaylight.org/gerrit/#/c/82381/ https://git.opendaylight.org/gerrit/#/c/82382/ https://git.opendaylight.org/gerrit/#/c/82383/ https://git.opendaylight.org/gerrit/#/c/82409/ https://git.opendaylight.org/gerrit/#/c/82461/ https://git.opendaylight.org/gerrit/#/c/82462/ https://git.opendaylight.org/gerrit/#/c/82465/ https://git.opendaylight.org/gerrit/#/c/82466/ https://git.opendaylight.org/gerrit/#/c/82467/ https://git.opendaylight.org/gerrit/#/c/82468/ https://git.opendaylight.org/gerrit/#/c/82769/ https://git.opendaylight.org/gerrit/#/c/82770/ https://git.opendaylight.org/gerrit/#/c/82773/ To evaluate these ... the simplest is to place a workload on SR1 and current stable/neon and compare JVM heap composition, most notably on followers and/or after a node was rebooted. If you are using segmented journal, you should also see the median size of persisted messages drop off in the JMX-exposed statistics it provides. Regards, Robert
signature.asc
Description: OpenPGP digital signature
_______________________________________________ controller-dev mailing list controller-dev@lists.opendaylight.org https://lists.opendaylight.org/mailman/listinfo/controller-dev
