You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Kirill Gusakov (Jira)" <ji...@apache.org> on 2021/12/02 20:22:00 UTC
[jira] [Comment Edited] (IGNITE-15725) Need to define scenarios related to data rebalancing and identifies possible corner cases
[ https://issues.apache.org/jira/browse/IGNITE-15725?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17452587#comment-17452587 ]
Kirill Gusakov edited comment on IGNITE-15725 at 12/2/21, 8:21 PM:
-------------------------------------------------------------------
As a first phase we decide to satisfy 2 cases:
* Direct change on baseline topology https://issues.apache.org/jira/browse/IGNITE-16010
* Change of replicas' number for the table https://issues.apache.org/jira/browse/IGNITE-15554
Partition changes must be addressed under https://issues.apache.org/jira/browse/IGNITE-16016 (it has some additional complexity, because cross-group raft rebalance is involved).
Further details about algorithm details will be provided under https://issues.apache.org/jira/browse/IGNITE-16020 .
was (Author: kgusakov):
As a first phase we decide to satisfy 2 cases:
* Direct change on baseline topology https://issues.apache.org/jira/browse/IGNITE-16010
* Change of replicas' number for the table https://issues.apache.org/jira/browse/IGNITE-15554
Further details about algorithm details will be provided under https://issues.apache.org/jira/browse/IGNITE-16020 .
> Need to define scenarios related to data rebalancing and identifies possible corner cases
> -----------------------------------------------------------------------------------------
>
> Key: IGNITE-15725
> URL: https://issues.apache.org/jira/browse/IGNITE-15725
> Project: Ignite
> Issue Type: Task
> Reporter: Vyacheslav Koptilin
> Assignee: Kirill Gusakov
> Priority: Major
> Labels: ignite-3
>
> Need to define scenarios that may lead to data rebalancing. The obvious ones are joining nodes to the cluster and leaving it.
> Also, corner cases should be addressed. For example, how the cluster should behave when the replica factor is 1.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)