cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-6477) Materialized Views (was: Global Indexes)
Date Wed, 15 Jul 2015 14:39:09 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-6477?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14628151#comment-14628151
] 

Sylvain Lebresne commented on CASSANDRA-6477:
---------------------------------------------

bq. So you're saying that IS NOT NULL should treat TTL'd columns as if they were already null?

I'm not, that would be terribly confusing (I personally couldn't live with it). I'm saying
that if initially we only support the case when all MV PK columns have {{IS NOT NULL}}, then
in the presence of TTL, a row should disappear from the MV as soon as the first MV PK column
expires. So, as I said earlier, I'm pretty sure we can get by by TTLing the whole MV row with
the smallest TTL of a MV PK column (Carl hinted at some problem with that suggestion, but
I haven't understood his point. So we should discuss that objection but 'a priori' I think
that suggestion work).

However, I'm also saying that I don't know how to handle that expiring case as soon as we
don't have the {{IS NOT NULL}} condition on some of the column, because in that case what
we should do when a column expire is _replacing_ the existing MV entry by a new one that have
a different PK (one with a {{null}} for the newly expired column), and that's hard to do.

> Materialized Views (was: Global Indexes)
> ----------------------------------------
>
>                 Key: CASSANDRA-6477
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6477
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: API, Core
>            Reporter: Jonathan Ellis
>            Assignee: Carl Yeksigian
>              Labels: cql
>             Fix For: 3.0 beta 1
>
>         Attachments: test-view-data.sh, users.yaml
>
>
> Local indexes are suitable for low-cardinality data, where spreading the index across
the cluster is a Good Thing.  However, for high-cardinality data, local indexes require querying
most nodes in the cluster even if only a handful of rows is returned.



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

Mime
View raw message