| From: | Jeff Cohen <jcohen(at)greenplum(dot)com> |
|---|---|
| To: | "Warren Turkal" <turkal(at)google(dot)com> |
| Cc: | "Ron Mayer" <rm_pg(at)cheapcomplexdevices(dot)com>, "Gavin Sherry" <swm(at)alcove(dot)com(dot)au>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Declarative partitioning grammar |
| Date: | 2008-01-12 00:39:19 |
| Message-ID: | B3513121-1093-45E6-94F9-AA9986B97190@greenplum.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Jan 11, 2008, at 4:16 PM, Warren Turkal wrote:
> On Jan 11, 2008 3:42 PM, Ron Mayer <rm_pg(at)cheapcomplexdevices(dot)com>
> wrote:
>> What would be the drawbacks of
>> CREATE TABLE tablename(...)
>> PARTITION BY function_taking_row_returning_partition_name
>> instead of the explicit types?
>
> Would that still allow the optimizer to work as well as it could? It
> seems that an arbitrary map like that can't be optimized very well as
> it might be too general.
We did look at allowing general functions for partitioning and this
was one concern. The other is that we want to enforce that a row
only gets inserted into a single partition, so we wanted a
declarative syntax where it was relatively easy to check that range
and list specifications don't overlap.
kind regards,
Jeff Cohen
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Merlin Moncure | 2008-01-12 00:46:36 | Re: Declarative partitioning grammar |
| Previous Message | Tom Lane | 2008-01-12 00:23:10 | Re: Transaction Snapshot Cloning |