[jira] [Commented] (ASTERIXDB-3389) Allow caching and evicting pages of columnar collection in cloud deployment

["ASF subversion and git services (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ASTERIXDB-3389?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17842176#comment-17842176 ] 

ASF subversion and git services commented on ASTERIXDB-3389:
------------------------------------------------------------

Commit ebde8c9f20297b8c958c33457ad64d3bbfcff722 in asterixdb's branch refs/heads/master from Wail Alkowaileet
[ https://gitbox.apache.org/repos/asf?p=asterixdb.git;h=ebde8c9f20 ]

[ASTERIXDB-3389][STO] Prep. columnar datasets for disk caching

- user model changes: no
- storage format changes: yes
- interface changes: yes

Details:
- Several modifications in columnar APIs to accommodate
  disk caching
- Pass read/write buffer cache context to LSMColumnBTree
  bulkloader and cursors
- Some refactoring in hyracks-cloud API (ASTERIXDB-3375)

Storage changes:
- Store the length (in bytes) of the mega leaf node in
  Page0 instead of the of the number of pages

Change-Id: Iababcc5fc1d4e5e2de36f9b26c3f86ffabfb4e54
Reviewed-on: https://asterix-gerrit.ics.uci.edu/c/asterixdb/+/18255
Integration-Tests: Jenkins <je...@fulliautomatix.ics.uci.edu>
Tested-by: Jenkins <je...@fulliautomatix.ics.uci.edu>
Reviewed-by: Wail Alkowaileet <wa...@gmail.com>
Reviewed-by: Murtadha Hubail <mh...@apache.org>


> Allow caching and evicting pages of columnar collection in cloud deployment
> ---------------------------------------------------------------------------
>
>                 Key: ASTERIXDB-3389
>                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-3389
>             Project: Apache AsterixDB
>          Issue Type: Improvement
>          Components: STO - Storage
>    Affects Versions: 0.9.10
>            Reporter: Wail Y. Alkowaileet
>            Assignee: Wail Y. Alkowaileet
>            Priority: Major
>              Labels: triaged
>             Fix For: 0.9.10
>
>
> In a cloud deployment, columnar collections will not be downloaded as we currently do. Instead, only pages of the requested columns will be brought up from the cloud. On disk pressure, those pages can be evicted (given some eviction policy/criteria).



--
This message was sent by Atlassian Jira
(v8.20.10#820010)