Re: using custom scan nodes to prototype parallel sequential scan

On Wed, Nov 12, 2014 at 1:24 PM, David Rowley <dgrowleyml(at)gmail(dot)com> wrote:

>
> On Tue, Nov 11, 2014 at 9:29 PM, Simon Riggs <simon(at)2ndquadrant(dot)com>
> wrote:
>
>>
>> This plan type is widely used in reporting queries, so will hit the
>> mainline of BI applications and many Mat View creations.
>> This will allow SELECT count(*) FROM foo to go faster also.
>>
>>
> We'd also need to add some infrastructure to merge aggregate states
> together for this to work properly. This means that could also work for
> avg() and stddev etc. For max() and min() the merge functions would likely
> just be the same as the transition functions.
>
>
It might make sense to make a new planner operator which can be responsible
for pulling from each of the individual parallel Agg nodes and then
aggregating over the results.

A couple of things that might be worth considering are whether we want to
enforce using parallel aggregation or let planner decide if it wants to do
a parallel aggregate or go with a single plan. For eg, the average
estimated size of groups might be one thing that planner may consider while
deciding between a parallel and a single execution plan.

I dont see merging states as an easy problem, and should perhaps be tackled
apart from this thread.

Also, do we want to allow parallelism only with GroupAggs?

Regards,

Atri