Mailing-List: contact dev-help@phoenix.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@phoenix.apache.org
Date: Thu, 5 Jan 2017 20:17:58 +0000 (UTC)
From: "Samarth Jain (JIRA)" <jira@apache.org>
To: dev@phoenix.apache.org
Message-ID: <JIRA.12927183.1451973975000.654275.1483647478444@Atlassian.JIRA>
In-Reply-To: <JIRA.12927183.1451973975000@Atlassian.JIRA>
References: <JIRA.12927183.1451973975000@Atlassian.JIRA> <JIRA.12927183.1451973975640@arcas>
Subject: [jira] [Commented] (PHOENIX-2565) Store data for immutable tables
 in single KeyValue
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Thu, 05 Jan 2017 20:18:05 -0000


    [ https://issues.apache.org/jira/browse/PHOENIX-2565?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15802411#comment-15802411 ] 

Samarth Jain commented on PHOENIX-2565:
---------------------------------------

If we go by the bitset approach where we don't store offset for missing or null columns, then figuring out where the bytes for a column are stored in the packed cell would no longer be a O(1) operation. For finding every column, on an average, we would have to go through half of the bitset to figure out the bytes for a column. We could possibly optimize that though by caching the bitset/offset information. This would come in handy especially when we have to evaluate multiple ArrayColumnExpressions against the packed cell. The caching would have to be done at a higher level though (parent of ArrayColumnExpression?). The cache would be built when evaluating the first ArrayColumnExpression. Subsequent evaluate() calls won't then have to compute where the bytes are.

> Store data for immutable tables in single KeyValue
> --------------------------------------------------
>
>                 Key: PHOENIX-2565
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-2565
>             Project: Phoenix
>          Issue Type: Improvement
>            Reporter: James Taylor
>            Assignee: Thomas D'Silva
>         Attachments: PHOENIX-2565-v2.patch, PHOENIX-2565-wip.patch, PHOENIX-2565.patch
>
>
> Since an immutable table (i.e. declared with IMMUTABLE_ROWS=true) will never update a column value, it'd be more efficient to store all column values for a row in a single KeyValue. We could use the existing format we have for variable length arrays.
> For backward compatibility, we'd need to support the current mechanism. Also, you'd no longer be allowed to transition an existing table to/from being immutable. I think the best approach would be to introduce a new IMMUTABLE keyword and use it like this:
> {code}
> CREATE IMMUTABLE TABLE ...
> {code}


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