[jboss-jira] [JBoss JIRA] Commented: (JGRP-1181) PING messages to entire cluster limit scalability.

[Bela Ban (JIRA)]

    [ https://jira.jboss.org/browse/JGRP-1181?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12551523#action_12551523 ] 

Bela Ban commented on JGRP-1181:
--------------------------------

One issue with only A, B and C responding is that we won't have physical addresses and logical names for D, E, F and G ! This means that, when we want to send a unicast message to G for instance, we have to fetch this first. This does NOT lead to incorrect behavior, but I suggest to set return_entire_cache="true"

> PING messages to entire cluster limit scalability.
> --------------------------------------------------
>
>                 Key: JGRP-1181
>                 URL: https://jira.jboss.org/browse/JGRP-1181
>             Project: JGroups
>          Issue Type: Feature Request
>            Reporter: Jonathan Oddy
>            Assignee: Bela Ban
>             Fix For: 2.11
>
>
> PING messages generate a response from every node that receives them. In a large cluster this will lead to a large number of responses needing to be processed. Making PINGs target-able (e.g. ping to only coordinators, ping to anyone with a view not matching ViewID N) could significantly reduce the number of nodes responding to a request.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira