You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Andrew Mashenkov (JIRA)" <ji...@apache.org> on 2017/01/27 12:40:24 UTC

[jira] [Updated] (IGNITE-4624) ScanQuery shows poor performance.

     [ https://issues.apache.org/jira/browse/IGNITE-4624?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Andrew Mashenkov updated IGNITE-4624:
-------------------------------------
    Attachment: Change_keyIterator_to_entryIterator_for_local_scan_squery.patch

I've make a simple fix and get a 30% speed up with iterating over partitions with ScanQuery.

Fix is not complete as it does not affect OffHeap cache and not bothered  with transactions.

> ScanQuery shows poor performance.
> ---------------------------------
>
>                 Key: IGNITE-4624
>                 URL: https://issues.apache.org/jira/browse/IGNITE-4624
>             Project: Ignite
>          Issue Type: Improvement
>          Components: cache
>    Affects Versions: 1.8
>            Reporter: Andrew Mashenkov
>         Attachments: Change_keyIterator_to_entryIterator_for_local_scan_squery.patch
>
>
> I've found that ScanQuery use cache KeySet iterator which make cache peek\entry calls for getting value.
> It is look like we can avoid this calls by using EntrySet iterator for iterating over local partitions.
> Start point is GridCacheQueryManager.onheapIterator(). 



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