accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Christopher Tubbs (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-164) Add support for wildcards/regexes in locality group setting.
Date Thu, 07 Feb 2013 20:01:13 GMT


Christopher Tubbs commented on ACCUMULO-164:

[~elserj] said:
I'm not sure how much I like the idea of tying some external dependency to the core implementation,
but having a very clear interface may reduce ambiguity. Keeping a locality group defined as
a set of column families helps, where a prefix match on column families defines a set of column
families, and thus locality group.

By "external dependency", I'm assuming you're talking about the suggestion to use JEXL. I
suppose I threw that out there carelessly, as an example for doing this with configuration,
rather than a pluggable implementation. After thinking about this some more, I think a pluggable
implementation for a "LocalityGrouper" interface to bin column families into locality groups,
would be preferable. The default implementation could be a "DiscreteSetGrouper" that reads
the existing per-table properties for backwards compatibility. If this doesn't add too much
complexity to the per-table configuration, it could be quite useful. There is a downside to
this, though. The RFile needs to know how families were binned when the file was written,
and I'm not sure they could do this easily with a pluggable mechanism, unless the grouper
implementation was always available at scan time. That gets tricky.

Yes, [~kturner], thanks for the clarification.
> Add support for wildcards/regexes in locality group setting.
> ------------------------------------------------------------
>                 Key: ACCUMULO-164
>                 URL:
>             Project: Accumulo
>          Issue Type: Improvement
>          Components: client, master, tserver
>            Reporter: John Vines
>             Fix For: 1.6.0
> We should look into adding the ability to specify locality group columns as either wildcarding
or regexes. I'm unsure of the feasibility of this, hence the lack of fix date.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message