You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Valentin Kulichenko (JIRA)" <ji...@apache.org> on 2016/11/21 23:48:58 UTC

[jira] [Created] (IGNITE-4256) Support closure broadcast with locked cache partitions

Valentin Kulichenko created IGNITE-4256:
-------------------------------------------

             Summary: Support closure broadcast with locked cache partitions
                 Key: IGNITE-4256
                 URL: https://issues.apache.org/jira/browse/IGNITE-4256
             Project: Ignite
          Issue Type: New Feature
          Components: general
    Affects Versions: 1.7
            Reporter: Valentin Kulichenko


This seems to be a feature that users are looking for in Ignite. Basically, for some cases it is useful to broadcast code that will scan local data (or query it using SQL) and process it somehow, potentially saving the result in another cache. This is mostly supported in Ignite - we have compute grid and local queries (both scan and SQL). The only issue is that in case of topology change the data can be moved and the result can be wrong.

We already support partition locking when executing {{affinityRun}} and {{affinityCall}}. So it seems to me that we should simply add a special mode to {{broadcast}} method that will lock all available partitions for specified cache(s).

API wise, one of the ways to do this is to add {{IgniteCompute.withNoDataMovement(String cacheNames...)}} method.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)