From: | Harald Fuchs <hf0923x(at)protecting(dot)net> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Using multi-row technique with COPY |
Date: | 2005-11-29 12:35:09 |
Message-ID: | 874q5vljia.fsf@srv.protecting.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
In article <20051129004435(dot)GR78939(at)pervasive(dot)com>,
"Jim C. Nasby" <jnasby(at)pervasive(dot)com> writes:
> On Sun, Nov 27, 2005 at 07:44:55PM +0000, Simon Riggs wrote:
>> not have any unique indexes or row triggers. It should be possible to
>> take advantage of this automatically when those requirements are met,
>> without any new options. Just as it was with Seq Scans, this is worth
>> about 10% reduction in CPU for a COPY FROM.
> <snip>
>> FSM access would need to change slightly to allow for whole-block-only
>> requests to be made for heaps, without damaging the average row length
>> calculation. It might be simpler to ignore FSM entirely?
> Does that mean that this fast copy would end up not re-using space on
> pages that have space available? ISTM that's something users would want
> to be able to over-ride. In fact, it seems like it shouldn't be a
> default behavior...
Why not? If you later do INSERT/UPDATE/DELETE, space will get reused
anyway, and if you don't (i.e. one-time bulk load on a constant
table), you should afterwards do a VACUUM FULL ANALYZE anyway.
From | Date | Subject | |
---|---|---|---|
Next Message | Zeugswetter Andreas DCP SD | 2005-11-29 14:27:59 | Re: gprof SELECT COUNT(*) results |
Previous Message | Martijn van Oosterhout | 2005-11-29 11:02:22 | Re: ice-broker scan thread |