Re: asynchronous and vectorized execution

On Wed, May 11, 2016 at 11:49 AM, Andres Freund <andres@anarazel.de> wrote:
> On 2016-05-11 10:12:26 -0400, Robert Haas wrote:
>> > I've to admit I'm not that convinced about the speedups in the !fdw
>> > case. There seems to be a lot easier avenues for performance
>> > improvements.
>>
>> What I'm talking about is a query like this:
>>
>> SELECT * FROM inheritance_tree_of_foreign_tables WHERE very_rarely_true;
>
> Note that I said "!fdw case".

Oh, wow, I totally missed that exclamation point.

>> > FWIW, I've even hacked something up for a bunch of simple queries, and
>> > the performance improvements were significant.  Besides it only being a
>> > weekend hack project, the big thing I got stuck on was considering how
>> > to exactly determine when to batch and not to batch.
>>
>> Yeah.  I think we need a system for signalling nodes as to when they
>> will be run to completion.  But a Boolean is somehow unsatisfying;
>> LIMIT 1000000 is more like no LIMIT than it it is like LIMIT 1.  I'm
>> tempted to add a numTuples field to every ExecutorState and give upper
>> nodes some way to set it, as a hint.
>
> I was wondering whether we should hand down TupleVectorStates to lower
> nodes, and their size determines the max batch size...

There's some appeal to that, but it seems complicated to make work.

>> >> Some care is required here because any
>> >> functions we execute as scan keys are run with the buffer locked, so
>> >> we had better not run anything very complicated.  But doing this for
>> >> simple things like integer equality operators seems like it could save
>> >> quite a few buffer lock/unlock cycles and some other executor overhead
>> >> as well.
>> >
>> > Hm. Do we really have to keep the page locked in the page-at-a-time
>> > mode? Shouldn't the pin suffice?
>>
>> I think we need a lock to examine MVCC visibility information.  A pin
>> is enough to prevent a tuple from being removed, but not from having
>> its xmax and cmax overwritten at almost but not quite exactly the same
>> time.
>
> We already batch visibility lookups in page-at-a-time
> mode. Cf. heapgetpage() / scan->rs_vistuples. So we can evaluate quals
> after releasing the lock, but before the pin is released, without that
> much effort.  IIRC that isn't used for index lookups, but that's
> probably a good idea.

The trouble with that is that if you fail the qual, you have to relock
the page.  Which kinda sucks, if the qual is really simple.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company