Thread: query has huge variance in execution times

postgres 8.3.5 on RHEL4 update 6

This query starts executing at 18:41:

cemdb=> select query_start,current_query from pg_stat_activity where
procpid=10022;           query_start          |

          current_query

-------------------------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  2010-03-30 18:41:11.685261-07 | select b.ts_id from
ts_stats_tranunit_user_daily b, ts_stats_tranunit_user_interval c where
b.ts_transet_incarnation_id = c.ts_transet_incarnation_id and
b.ts_tranunit_id = c.ts_tranunit_id and b.ts_user_incarnation_id =
c.ts_user_incarnation_id and c.ts_interval_start_time >= $1 and
c.ts_interval_start_time < $2 and b.ts_interval_start_time >= $3 and
b.ts_interval_start_time < $4
(1 row)

about 5 mins later, I, suspecting problems, do (the values are the same
as for $1 et al above; EXPLAIN was done on purpose to keep stats
[hopefully] the same as when pid 10022 started; there are 80,000 rows in
each of the 2 tables at the time of this EXPLAIN and when 10022 started):

cemdb=> explain select b.ts_id from ts_stats_tranunit_user_daily b,
ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= '2010-3-29 01:00' and
c.ts_interval_start_time < '2010-3-29 02:00' and
b.ts_interval_start_time >= '2010-3-29' and b.ts_interval_start_time <
'2010-3-30';

                    QUERY PLAN

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  Merge Join  (cost=33574.89..34369.38 rows=25207 width=8)
    Merge Cond: ((b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id) AND (b.ts_tranunit_id = c.ts_tranunit_id)
AND (b.ts_user_incarnation_id = c.ts_user_incarnation_id))
    ->  Sort  (cost=13756.68..13854.96 rows=78623 width=32)
          Sort Key: b.ts_transet_incarnation_id, b.ts_tranunit_id,
b.ts_user_incarnation_id
          ->  Index Scan using ts_stats_tranunit_user_daily_starttime on
ts_stats_tranunit_user_daily b  (cost=0.00..10560.13 rows=78623 width=32)
                Index Cond: ((ts_interval_start_time >= '2010-03-29
00:00:00-07'::timestamp with time zone) AND (ts_interval_start_time <
'2010-03-30 00:00:00-07'::timestamp with time zone))
    ->  Sort  (cost=19818.21..19959.72 rows=113207 width=24)
          Sort Key: c.ts_transet_incarnation_id, c.ts_tranunit_id,
c.ts_user_incarnation_id
          ->  Index Scan using ts_stats_tranunit_user_interval_starttime
on ts_stats_tranunit_user_interval c  (cost=0.00..15066.74 rows=113207
width=24)
                Index Cond: ((ts_interval_start_time >= '2010-03-29
01:00:00-07'::timestamp with time zone) AND (ts_interval_start_time <
'2010-03-29 02:00:00-07'::timestamp with time zone))
(10 rows)

cemdb=> \q

I then run the query manually:

[root@rdl64xeoserv01 log]# time PGPASSWORD=quality psql -U admin -d
cemdb -c "select b.ts_id from ts_stats_tranunit_user_daily b,
ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= '2010-3-29 01:00' and
c.ts_interval_start_time < '2010-3-29 02:00' and
b.ts_interval_start_time >= '2010-3-29' and b.ts_interval_start_time <
'2010-3-30'" > /tmp/select.txt 2>&1

real    0m0.813s
user    0m0.116s
sys     0m0.013s

I let process 10022 run for an hour. an strace shows lots of I/O:

[root@rdl64xeoserv01 log]# strace -p 10022
read(18, "\214\2\0\0\374<\200#\1\0\0\0<\0P\3\0 \4 \0\0\0\0\320\234"...,
8192) = 8192
semop(73007122, 0xbfe0fc20, 1)          = 0
_llseek(18, 538451968, [538451968], SEEK_SET) = 0
read(18, "\214\2\0\0\274\347\t#\1\0\0\0<\0P\3\0 \4 \0\0\0\0\320\234"...,
8192) = 8192
_llseek(18, 535928832, [535928832], SEEK_SET) = 0
read(18, "\214\2\0\0\310\300\226\"\1\0\0\0<\0P\3\0 \4 \0\0\0\0\320"...,
8192) = 8192
_llseek(18, 532398080, [532398080], SEEK_SET) = 0

<many more similar lines>

I then kill 10022 and the application retries the same query:

[10022-cemdb-admin-2010-03-30 19:02:37.460 PDT]FATAL:  terminating
connection due to administrator command
[10022-cemdb-admin-2010-03-30 19:02:37.460 PDT]STATEMENT:  select
b.ts_id from ts_stats_tranunit_user_daily b,
ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= $1 and c.ts_interval_start_time < $2 and
b.ts_interval_start_time >= $3 and b.ts_interval_start_time < $4

[10820-cemdb-admin-2010-03-30 19:02:40.363 PDT]LOG:  duration: 1096.598
ms  execute <unnamed>: select b.ts_id from ts_stats_tranunit_user_daily
b, ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= $1 and c.ts_interval_start_time < $2 and
b.ts_interval_start_time >= $3 and b.ts_interval_start_time < $4

Ideas on why the big difference in execution times of the same query and
how to avoid same are solicited. I tend to doubt, but don't know how to
prove, that the query stats for these 2 tables were updated between the
start of pid 10022 and when I did the EXPLAIN.

Thanks,
Brian

--
- David T. Wilson
david.t.wilson@gmail.com

On 03/31/2010 12:37 AM, David Wilson [david.t.wilson@gmail.com] wrote:
> These won't necessarily get the same plan. If you want to see what plan
> the prepared query is getting, you'll need to prepare it ("prepare foo
> as <query>") and then explain *that* via "explain execute foo".
>
> The prepared version likely has a much more generic plan, whereas the
> regular query gets optimized for the actual values provided.

I didn't know this. Thanks. The plans are indeed different:

cemdb=# prepare sq as select b.ts_id from ts_stats_tranunit_user_daily
b, ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= $1 and c.ts_interval_start_time < $2 and
b.ts_interval_start_time >= $3 and b.ts_interval_start_time < $4;
cemdb=# explain execute sq('2010-3-29 01:00', '2010-3-29 02:00',
'2010-3-29', '2010-3-30');
                                       QUERY PLAN

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  Merge Join  (cost=7885.37..8085.91 rows=30 width=8)
    Merge Cond: ((b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id) AND (b.ts_tranunit_id = c.ts_tranunit_id)
AND (b.ts_user_incarnation_id = c.ts_user_incarnation_id))
    ->  Sort  (cost=1711.82..1716.81 rows=3994 width=32)
          Sort Key: b.ts_transet_incarnation_id, b.ts_tranunit_id,
b.ts_user_incarnation_id
          ->  Index Scan using ts_stats_tranunit_user_daily_starttime on
ts_stats_tranunit_user_daily b  (cost=0.00..1592.36 rows=3994 width=32)
                Index Cond: ((ts_interval_start_time >= $3) AND
(ts_interval_start_time < $4))

cemdb=# explain select b.ts_id from ts_stats_tranunit_user_daily b,
ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= '2010-3-29 01:00' and
c.ts_interval_start_time < '2010-3-29 02:00' and
b.ts_interval_start_time >= '2010-3-29' and b.ts_interval_start_time <
'2010-3-30';

                    QUERY PLAN

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  Hash Join  (cost=291965.90..335021.46 rows=13146 width=8)
    Hash Cond: ((c.ts_transet_incarnation_id =
b.ts_transet_incarnation_id) AND (c.ts_tranunit_id = b.ts_tranunit_id)
AND (c.ts_user_incarnation_id = b.ts_user_incarnation_id))
    ->  Index Scan using ts_stats_tranunit_user_interval_starttime on
ts_stats_tranunit_user_interval c  (cost=0.00..11783.36 rows=88529 width=24)
          Index Cond: ((ts_interval_start_time >= '2010-03-29
01:00:00-07'::timestamp with time zone) AND (ts_interval_start_time <
'2010-03-29 02:00:00-07'::timestamp with time zone))
    ->  Hash  (cost=285681.32..285681.32 rows=718238 width=32)
          ->  Index Scan using ts_stats_tranunit_user_daily_starttime on
ts_stats_tranunit_user_daily b  (cost=0.00..285681.32 rows=718238 width=32)
                Index Cond: ((ts_interval_start_time >= '2010-03-29
00:00:00-07'::timestamp with time zone) AND (ts_interval_start_time <
'2010-03-30 00:00:00-07'::timestamp with time zone))
(7 rows)

    ->  Sort  (cost=6173.55..6218.65 rows=36085 width=24)
          Sort Key: c.ts_transet_incarnation_id, c.ts_tranunit_id,
c.ts_user_incarnation_id
          ->  Index Scan using ts_stats_tranunit_user_interval_starttime
on ts_stats_tranunit_user_interval c  (cost=0.00..4807.81 rows=36085
width=24)
                Index Cond: ((ts_interval_start_time >= $1) AND
(ts_interval_start_time < $2))
(10 rows)

I notice that the row estimates are substantially different; this is due
to the lack of actual values?

But, this prepared query runs in ~4 secs:

[root@rdl64xeoserv01 log]# cat /tmp/select.sql
prepare sq as select b.ts_id from ts_stats_tranunit_user_daily b,
ts_stats_tranunit_user_interval c where b.ts_transet_incarnation_id =
c.ts_transet_incarnation_id and b.ts_tranunit_id = c.ts_tranunit_id and
b.ts_user_incarnation_id = c.ts_user_incarnation_id and
c.ts_interval_start_time >= $1 and c.ts_interval_start_time < $2 and
b.ts_interval_start_time >= $3 and b.ts_interval_start_time < $4;
execute sq('2010-3-29 01:00', '2010-3-29 02:00', '2010-3-29', '2010-3-30

[root@rdl64xeoserv01 log]# time PGPASSWORD=quality psql -U postgres -d
cemdb -f /tmp/select.sql  > /tmp/select1.txt 2>&1
real    0m4.131s
user    0m0.119s
sys     0m0.007s

so the question still remains: why did it take > 20 mins? To see if it
was due to autovacuum running ANALYZE, I turned off autovacuum, created
a table using SELECT * INTO temp FROM ts_stats_tranunit_user_daily,
added the index on ts_interval_start_time and ran the prepared query
with temp, but the query completed in a few secs.

Brian

It's possible that statistics were updated between the >20 minute run and your most recent prepared query test. In fact, comparing the plans between your two emails, it's quite likely, as even the non-prepared versions are not producing the same plan or the same estimates; it's therefore possible that your problem has already corrected itself if you're unable to duplicate the 20 minute behaviour at this point.

Taking a look at the statistics accuracy with an explain analyze might still be informative, however.

--
- David T. Wilson
david.t.wilson@gmail.com