phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ohad Shacham (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (PHOENIX-4278) Implement pure client side transactional index maintenance
Date Mon, 29 Jan 2018 16:08:00 GMT

    [ https://issues.apache.org/jira/browse/PHOENIX-4278?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16343550#comment-16343550
] 

Ohad Shacham commented on PHOENIX-4278:
---------------------------------------

As the bot stated, I uploaded a pull request for this Jira. 

[~jamestaylor], just wanted to note that in this commit I also changed the way deletion is
being made. Instead of using Tephra's deletion marker I am using Tephra's regular deletion
api. This will be crucial for Omid, since using Omid's api will let Omid know that a deletion
has been made and that update of its shadow cell upon commit is needed. Update the deletion
marker directly in HBase will result in removing the commit information from the commit table,
without updating the shadow cells, and the index update will be treated as not committed.

> Implement pure client side transactional index maintenance
> ----------------------------------------------------------
>
>                 Key: PHOENIX-4278
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-4278
>             Project: Phoenix
>          Issue Type: Improvement
>            Reporter: James Taylor
>            Assignee: Ohad Shacham
>            Priority: Major
>
> The index maintenance for transactions follows the same model as non transactional tables
- coprocessor based on data table updates that looks up previous row value to perform maintenance.
This is necessary for non transactional tables to ensure the rows are locked so that a consistent
view may be obtained. However, for transactional tables, the time stamp oracle ensures uniqueness
of time stamps (via transaction IDs) and the filtering handles a scan seeing the "true" last
committed value for a row. Thus, there's no hard dependency to perform this on the server
side.
> Moving the index maintenance to the client side would prevent any RS->RS RPC calls
(which have proved to be troublesome for HBase). It would require returning more data to the
client (i.e. the prior row value), but this seems like a reasonable tradeoff.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message