[
https://issues.apache.org/jira/browse/PHOENIX-3148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15431927#comment-15431927
]
James Taylor commented on PHOENIX-3148:
---------------------------------------
Thanks for the patch and the perf analysis. Looks very good. Here's some
feedback:
- How about we name the property {{phoenix.table.client.cache.encoding}} with a
string value (for now) of {{protobuf}} and {{object}}? We can have an enum that
captures the possible values. I think we can make the default {{object}} for
4.8.1 to err on the conservative side (and pending a Pherf run with the default
as {{protobuf}}). If Pherf looks fine for existing workload, we can switch to
{{protobuf}} as the default.
- For the PTableImpl equals and hashCode(), I think you need to take into
account the tenantId (which is part of the PTableKey). I think you can just
implement these methods by delegating them to {{key}}.
- Maybe the PTableRefFactory.getFactory() method can just get passed the
QueryServices and decide which factory to use based on the properties?
> Reduce size of PTable so that more tables can be cached in the metada cache.
> ----------------------------------------------------------------------------
>
> Key: PHOENIX-3148
> URL: https://issues.apache.org/jira/browse/PHOENIX-3148
> Project: Phoenix
> Issue Type: Bug
> Reporter: Thomas D'Silva
> Assignee: Thomas D'Silva
> Labels: argus
> Fix For: 4.8.1
>
> Attachments: PHOENIX-3148-v2.patch, PHOENIX-3148-v3.patch,
> PHOENIX-3148.patch, comparison.png
>
>
> According to PHOENIX-2995, the current size is 7KB per PTable which works out
> to enabling 140K PTables per 1GB.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
