[
https://issues.apache.org/jira/browse/CASSANDRA-6477?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14626414#comment-14626414
]
Sylvain Lebresne commented on CASSANDRA-6477:
---------------------------------------------
bq. NULL, however, is not something people think about since it's the default
value implicitly for Cassandra.
Again, I'm not trying to say that it's not something we should consider. I'm
mainly saying that imo we should do CASSANDRA-9796 anyway so we don't have the
internal limitation. Maybe we can impose limitations for the sake of protecting
against foot-shooting, like requiring that at least one column of the MV
partition key is a primary key column of the base table (whithout requiring all
of them to be).
bq. If you are writing to a table with RF=3 and only one replica has a value
but the other two do not then we will make one MV replica have a defined
partition key and the other two replicas will have a NULL
Not sure I follow. Nulls should just be "just another value" and there should
really be any special casing for them. How is what you say different from
writing a new value at RF=3 and only one replica getting the new value?
> 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)
