Message-ID: <4EA6952E.90209@comodo.com>
Date: Tue, 25 Oct 2011 16:23:34 +0530
From: Robins Tharakan <robins.tharakan@comodo.com>
Organization: Comodo
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
	rv:1.9.2.23) Gecko/20110922 Thunderbird/3.1.15
MIME-Version: 1.0
To: pgsql-performance@postgresql.org
Subject: Re: Bad plan by Planner (Already resolved?)
References: <4E9D2ECA0200002500042131@gw.wicourts.gov>
In-Reply-To: <4E9D2ECA0200002500042131@gw.wicourts.gov>
Content-Type: multipart/signed; protocol="application/pkcs7-signature";
	micalg=sha1; boundary="------------ms000205090001090805060403"

This is a cryptographically signed message in MIME format.

--------------ms000205090001090805060403
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: quoted-printable

Thanks Kevin,

That's a pretty neat way to managing (at least) minor upgrades. Like I=20
said, this place is new, and so although I'm quite positive about=20
upgrading to the latest, I should probably take things one-at-a-time and =

bring in this idea of implementing regular updates sometime in the future=
=2E

As for the query, I tried the same query on an alternate machine, and=20
this is how EXPLAIN ANALYZE came up. Its much faster than the earlier=20
slow query, but nowhere near the performance of the second query shown=20
earlier. Do I have to live with that until this is implemented (if I am=20
only doing a minor version upgrade) or am I missing something else here?

I've provided the EXPLAIN ANALYZE as well as the web-link for a pretty=20
output of the EXPLAIN ANALYZE for your review.


ORIGINAL QUERY (on PostgreSQL 8.4.9):
http://explain.depesz.com/s/bTm

EXPLAIN ANALYZE SELECT field_a FROM large_table_a JOIN large_table_b=20
USING (field_a) WHERE field_b IN (SELECT large_table_b.field_b FROM=20
large_table_b WHERE field_a =3D 2673056) ;

-------------------------------------------------------------------------=
-------------------------------------------------------------------------=
----------
  Nested Loop  (cost=3D132.97..194243.54 rows=3D156031 width=3D4) (actual=
=20
time=3D6.612..43179.524 rows=3D2120 loops=3D1)
    ->  Nested Loop  (cost=3D132.97..1107.63 rows=3D156031 width=3D4) (ac=
tual=20
time=3D6.576..29122.017 rows=3D6938 loops=3D1)
          ->  HashAggregate  (cost=3D132.97..133.96 rows=3D99 width=3D4) =

(actual time=3D6.543..12.726 rows=3D2173 loops=3D1)
                ->  Index Scan using "IX_large_table_b_SigId" on=20
large_table_b  (cost=3D0.00..132.56 rows=3D164 width=3D4) (actual=20
time=3D0.029..3.425 rows=3D2173 loops=3D1)
                      Index Cond: (field_a =3D 2673056)
          ->  Index Scan using "IX_large_table_b_field_b" on=20
large_table_b  (cost=3D0.00..9.81 rows=3D2 width=3D8) (actual=20
time=3D6.732..13.384 rows=3D3 loops=3D2173)
                Index Cond: (public.large_table_b.field_b =3D=20
public.large_table_b.field_b)
    ->  Index Scan using "PK_large_table_a" on large_table_a=20
(cost=3D0.00..1.23 rows=3D1 width=3D4) (actual time=3D2.021..2.021 rows=3D=
0=20
loops=3D6938)
          Index Cond: (large_table_a.field_a =3D public.large_table_b.fie=
ld_a)
  Total runtime: 43182.975 ms


OPTIMIZED QUERY (on PostgreSQL 8.4.7):
http://explain.depesz.com/s/emO

EXPLAIN ANALYZE SELECT s1.field_a FROM large_table_a JOIN large_table_b=20
s1 USING (field_a) JOIN large_table_b s2 ON s1.field_b =3D s2.field_b=20
WHERE s2.field_a =3D 2673056;

-------------------------------------------------------------------------=
-------------------------------------------------------------------------=
----------
  Nested Loop  (cost=3D0.00..2356.98 rows=3D494 width=3D4) (actual=20
time=3D0.086..96.056 rows=3D2120 loops=3D1)
    ->  Nested Loop  (cost=3D0.00..1745.51 rows=3D494 width=3D4) (actual =

time=3D0.051..48.900 rows=3D6938 loops=3D1)
          ->  Index Scan using "IX_large_table_b_SigId" on large_table_b =

s2  (cost=3D0.00..132.56 rows=3D164 width=3D4) (actual time=3D0.028..3.41=
1=20
rows=3D2173 loops=3D1)
                Index Cond: (field_a =3D 2673056)
          ->  Index Scan using "IX_large_table_b_field_b" on=20
large_table_b s1  (cost=3D0.00..9.81 rows=3D2 width=3D8) (actual=20
time=3D0.007..0.012 rows=3D3 loops=3D2173)
                Index Cond: (s1.field_b =3D s2.field_b)
    ->  Index Scan using "PK_large_table_a" on large_table_a=20
(cost=3D0.00..1.23 rows=3D1 width=3D4) (actual time=3D0.004..0.004 rows=3D=
0=20
loops=3D6938)
          Index Cond: (large_table_a.field_a =3D s1.field_a)
  Total runtime: 98.165 ms


--
Robins Tharakan

On 10/18/2011 06:16 PM, Kevin Grittner wrote:
> We build from source, and we include the minor release number in the
> prefix for the build, so we can have both old and new software
> installed side-by-side.  The path for the client-side executables we
> do through a symbolic link, so we can switch that painlessly.  And we
> assign the prefix used for the server to an environment variable in
> our services script.  So here is our process:
>
>   - Build and install the new minor release.
>   - Change the symlink to use it for clients (like pg_dump and psql).
>   - Change the service script line that sets the prefix to point to
>     the new minor release.
>   - Run the service script with "stop" and then run the service script
>     with "start".  (Unless your service script does a restart by using
>     stop and start, don't run it with "restart", because a PostgreSQL
>     restart won't pick up the new executables.)
>
> There is literally no more down time than it takes to stop and start
> the database service.  Our client software retries on a broken
> connection, so we can even do this while users are running and they
> just get a clock for a few seconds; but we usually prefer not to
> cause even that much disruption, at least during normal business
> hours.  We have enough hardware to load balance off of one machine at
> a time to do this without interruption of service.
>
> There are sometimes bugs fixed in a minor release which require
> cleanup of possibly damaged data, like what I mentioned above.  You
> may need to vacuum or reindex something to recover from the damage
> caused by the now-fixed bug, but the alternative is to continue to
> run with the damage.  I don't understand why someone would knowingly
> choose that.
>
> Really, it is worthwhile to keep up on minor releases.
> -Kevin


--------------ms000205090001090805060403
Content-Type: application/pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms000205090001090805060403--