Re: WIP: Avoid creation of the free space map for small tables

On Sat, Dec 8, 2018 at 6:35 PM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
> On Fri, Dec 7, 2018 at 7:25 PM John Naylor <jcnaylor(at)gmail(dot)com> wrote:
> >
> > On 12/6/18, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> > > On Thu, Dec 6, 2018 at 10:53 PM John Naylor <jcnaylor(at)gmail(dot)com>
wrote:
> > >>
> > >> I've added an additional regression test for finding the right block

I did run some performance tests on the latest patch v11, I see small
regression in execution time of COPY statement. Tests I have used is same
as provided in [1] just that I ran it for fill factor 20 and 70. Here are
my results!

Machine : cthulhu (Intel based 8 numa machine)
Server setting is default, configured with HEAP_FSM_CREATION_THRESHOLD = 4,
Entire data directory was on HDD.

Results are execution time(unit ms) taken by copy statement when number of
records equal to exact number which fit HEAP_FSM_CREATION_THRESHOLD = 4
pages. For fill factor 20 it is till tid (3, 43) and for scale factor 70
till tid (3, 157). Result is taken as a median of 10 runs.

Fill factor 20
Tables Base Patch % of increase in execution time
500 121.97 125.315 2.7424776584
1000 246.592 253.789 2.9185861666

Fill factor 70
500 211.502 217.128 2.6600221275
1000 420.309 432.606 2.9257046601

So 2-3% consistent regression, And on every run I can see for patch v11
execution time is slightly more than base. I also tried to insert more
records till 8 pages and same regression is observed! So I guess even
HEAP_FSM_CREATION_THRESHOLD = 4 is not perfect!

[1]
https://www.postgresql.org/message-id/CAJVSVGX%3D2Q52fwijD9cjeq1UdiYGXns2_9WAPFf%3DE8cwbFCDvQ%40mail.gmail.com

--
Thanks and Regards
Mithun Chicklore Yogendra
EnterpriseDB: http://www.enterprisedb.com