[
https://issues.apache.org/jira/browse/PHOENIX-3547?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15781919#comment-15781919
]
James Taylor commented on PHOENIX-3547:
---------------------------------------
That's a good point. You'd need to rebuild all indexes on views. We've done
this in the past as an upgrade step that kicks off asynchronous jobs to rebuild
indexes as needed, but it's not ideal. Some users may depend on an index being
active all the time.
FWIW, you have 128K unique view indexes per tenant. Would it be possible to
explain your idea/use case a bit more to understand why this wouldn't be
enough? Are you envisioning many tenants creating the same index, one that
doesn't refer to any tenant-specific columns they've added? This type of index
would be better handled as a global index instead of an index on a view. If
it's an index on a tenant-specific view then it doesn't really make sense for
other tenants to use it because they don't have the column being indexes.
> Promote CATALOG.VIEW_INDEX_ID to an int
> ---------------------------------------
>
> Key: PHOENIX-3547
> URL: https://issues.apache.org/jira/browse/PHOENIX-3547
> Project: Phoenix
> Issue Type: New Feature
> Reporter: Jeremy Huffman
> Priority: Minor
>
> Increase the size of CATALOG.VIEW_INDEX_ID from smallint to int to support a
> large number of indexed views on a single table.
> Per James: "The code would just need to be tolerant when reading the data if
> the length is two byte short versus four byte int. At write time, we'd just
> always write an int."
> See:
> https://lists.apache.org/thread.html/22849e4fc73452cee3bea763cf6d5af7164dedcb44573ba6b9f452a2@%3Cuser.phoenix.apache.org%3E
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
