phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James Taylor (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (PHOENIX-4552) Allow ROW_TIMESTAMP declaration for non PK column
Date Thu, 25 Jan 2018 20:05:00 GMT

    [ https://issues.apache.org/jira/browse/PHOENIX-4552?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16339795#comment-16339795
] 

James Taylor commented on PHOENIX-4552:
---------------------------------------

Had one more thought, [~spetschu]: this optimization won't help your use case. Since you're
already handling schema drift in subsequent CSV loads, you'll always find a latest cell value
for every column in your schema. Thus earlier versions of column values will never been seen
by your queries. This optimization would only help if you didn't handle this case. For example,
if the schema varied greatly between CSV loads and the current load didn't mask column values
from the previous load.

> Allow ROW_TIMESTAMP declaration for non PK column
> -------------------------------------------------
>
>                 Key: PHOENIX-4552
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-4552
>             Project: Phoenix
>          Issue Type: New Feature
>            Reporter: Stephen Petschulat
>            Priority: Minor
>
> By declaring a ROW_TIMESTAMP constraint on a Phoenix table, it does two things 1) expose
the hbase native timestamp as this column and 2) prepend your primary key with this timestamp
as well.
> It would be useful to have a similar feature that only exposes the hbase native timestamp.
This would allow explicit setting of the timestamp when upserting data while allowing multiple
hbase versions. It is possible to then query for that specific key and version(s).
> Potential approach:
> {code:sql}
> CREATE TABLE COMMENTS (
>    COMMENT_ID INT NOT NULL,
>    REVISION_NUM BIGINT NOT NULL ROW_TIMESTAMP,    // NEW use of keyword
>    COMMENT_BODY TEXT
>    CONSTRAINT PK PRIMARY KEY(COMMENT_ID))
> UPSERT INTO COMMENTS (123, 1, 'edit 1 comment')
> UPSERT INTO COMMENTS (123, 2, 'edit 2 of comment')
> UPSERT INTO COMMENTS (123, 3, 'edit 3 of comment')
> {code}
>  
> Current behavior of ROW_TIMESTAMP would create a new primary for each upsert, so querying
by primary key is no longer straightforward when you don't know the version number at query
time. 
> {code:sql}
> SELECT * FROM COMMENTS WHERE COMMENT_ID = 123  // => returns most recent version
'edit 3 of comment'
> SELECT * FROM COMMENTS WHERE COMMENT_ID = 123 AND REVISION_NUM = 1   // => returns
explicit version 'edit 1 comment'
> {code}
>  
> It can also be useful to return multiple versions (related: PHOENIX-590)
> {code:sql}
> SELECT * FROM COMMENTS WHERE COMMENT_ID = 123 AND REVISION_NUM < 3   // => returns 2
rows
> {code}
>  
> Or just the highest version less than or equal to a particular version (allowing snapshot
queries):
> {code:sql}
> // set CurrentSCN=2 on connection
> SELECT * FROM COMMENTS WHERE COMMENT_ID = 123 // => returns 'edit 2 of comment'
> {code}
> CurrentSCN already allows this type of snapshot query but not against an explicitly set
timestamp with multiple versions. The primary key injection prevents this. The above query would
behave similar to:
> {code:java}
> scan 'COMMENTS', {TIMERANGE => [0, <maxversionid+1>]}
> {code}
>  This returns the highest versioned value for each key that is less than a specified
maximum version number.
>  
>  



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

Mime
View raw message