[
https://issues.apache.org/jira/browse/CASSANDRA-6477?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14363490#comment-14363490
]
Carl Yeksigian commented on CASSANDRA-6477:
-------------------------------------------
I pushed an updated branch
[here|https://github.com/carlyeks/cassandra/tree/ticket/6477-2]. From the last
list, the only remaining outstanding issue is collection cells. I'm inclined to
exclude those from the initial implementation.
[~m0nstermind]: these are all great suggestions, but we should add some of
these as follow on tickets. It makes sense to change to {{INCLUDE}}, so I've
replaced {{DENORMALIZED}} already, and I changed the order of the mutation
statements (good catch!).
Issues we should address in new tickets:
# Composite Indexes
# Changing index CF properties
# 2i and GI mixed mode
I'm still not convinced that you should have 2i and GI on the same column,
since they are for very different use cases, even in the transitional period
# For the initial build, I think there are probably a lot of ways to make it
faster than what I have, but this is just a first cut
An idea I've had is to create mutations from repaired sstables on a single
replica; for unrepaired data, all the replicas would need to create them
# Currently, base table evolution is not handled, and any such modification
will actually cause an IRE, but should be addressed
> 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
>
>
> 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)
