cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-9471) Columns should be backed by a BTree, not an array
Date Fri, 03 Jul 2015 10:16:05 GMT


Sylvain Lebresne commented on CASSANDRA-9471:

bq. for the normal worries of code atrophy

Yes, that is something to take into account. However, it's also a utility class, one that
is meant to be used a lot in the codebase. And the indexability code both already written.
So if it "doesn't introduce significant complexity", it does feels like a relatively good
deal. Basically, I would hate to spend more time pulling the already written functionality
out to maybe end up someday having a good use of this, but ending up doing something less
efficient just because it's not there. Besides, it's totally possible it will be used by {{Columns}}
in the end :)

Anyway, I don't want to sound insistent, it's not that I absolutely want it. Just offering
that maybe simply rebasing that ticket now would avoid pushing that to when we might be even
shorter on resources than we are, doesn't precludes considering better alternative for {{Columns}}
later, and won't waste all that much work if we do end up changing {{Columns}} but keep the
indexability as "generally useful". 

> Columns should be backed by a BTree, not an array
> -------------------------------------------------
>                 Key: CASSANDRA-9471
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Benedict
>            Assignee: Benedict
>             Fix For: 3.0 beta 1
> Follow up to 8099. 
> We have pretty terrible lookup performance as the number of columns grows (linear). In
at least one location, this results in quadratic performance. 
> We don't however want this structure to be either any more expensive to build, nor to
store. Some small modifications to BTree will permit it to serve here, by permitting efficient
lookup by index, and calculation _of_ index for a given key.

This message was sent by Atlassian JIRA

View raw message