You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ozone.apache.org by "Arpit Agarwal (Jira)" <ji...@apache.org> on 2020/06/02 17:49:00 UTC

[jira] [Updated] (HDDS-1528) Buffer Pool in BlockInputStream

     [ https://issues.apache.org/jira/browse/HDDS-1528?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Arpit Agarwal updated HDDS-1528:
--------------------------------
    Target Version/s: 0.7.0

> Buffer Pool in BlockInputStream
> -------------------------------
>
>                 Key: HDDS-1528
>                 URL: https://issues.apache.org/jira/browse/HDDS-1528
>             Project: Hadoop Distributed Data Store
>          Issue Type: Improvement
>            Reporter: Hanisha Koneru
>            Priority: Major
>
> In BlockInputStream, when a new chunk is being read, the data in old buffers is thrown away. So, if we want to read data in chunk1 and then chunk2 and then go back to reading chunk1, we would have to fetch chunk1 again from the DNs.
> Reads can be optimized if we maintain a buffer pool instead instead of only one buffer. This way, we can cache chunk data for re-reads.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: ozone-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: ozone-issues-help@hadoop.apache.org