Обсуждение: Bitmap Heap Scan anomaly

<div style="font-family:times new roman, new york, times, serif;font-size:12pt">I have done the following test and I am
unableto understand the results.  I have tried debugging the code and I have reached down to the Storage Layer.  I am
playingwith the optimizer etc.. I no very little about the internals of the Executor.<br /><br />If you could point out
tome what possible explanation for such anomaly I would be very glad.<br /><br />Thanks,<br />Makarona<br /><br />My
Test:<br/><br />Setup:<br />-------<br />I have created two very similar tables mycorr_10 and mycorr_100, attribute
namesare {key,a,b} for both tables. <br />I added 16 M rows in both tables in the following fashion:<br />    I gave a
randomvalue to each attribute key ( dont care )<br />    Values in a,b take a random value from [1-16M]<br />    In the
caseof mycorr_10 I set a random 10% of the a=b<br />    In the case of mycorr_100 I set all a=b<br />    I create
index{a,b}on both tables<br />    I VACUUM ANALYZE<br />p.s. I am trying to simulate an optimizer cardinality
estimationerror due to Independence assumption.<br /><br />Query :<br />SELECT count(key)<br />FROM 
mycorr_10                                             -- (or mycorr_100)<br />WHERE a>15900000 and b>15900000;<br
/><br/>Explain:<br />----------<br />As expected using the independence assumption the Planner chooses to use the index
forboth tables cases:<br /> Aggregate([4130.82][4130.83][1][94083.95][94083.96][1] width=4)<br />   ->  Bitmap Heap
Scanon mycorr_100([1997.92][4129.41][566][2021.57][93846.00][95177] width=4)<br />         Recheck Cond: ((a >
15900000)AND (b > 15900000))<br />         ->  Bitmap Index Scan on
ab_100([0.00][1997.77][566][0.00][1997.77][95177]width=0)<br />               Index Cond: ((a > 15900000) AND (b
>15900000))<br />(5 rows)<br /><br />p.s.<br />Explain output may seem weird as i have changes it a bit.<br /><br
/><br/>Explain Analyze<br />---------------------<br /><br />restart postgres<br />echo 1 >
/proc/sys/vm/drop_caches    (drop file system caches)<br />explain analyze select count(key) from mycorr_10 where
a>15900000and b>15900000;<br />restart postgres<br />echo 1 > /proc/sys/vm/drop_caches<br />explain analyze
selectcount(key) from mycorr_100 where a>15900000 and b>15900000;<br /><br /><br /><br />Result for
mycorr_100:<br/>---------------------------<br /> Aggregate([4130.82][4130.83][1][94083.95][94083.96][1] width=4)
(actualtime=11424.077..11424.078 rows=1 loops=1)<br />   ->  Bitmap Heap Scan on
mycorr_100([1997.92][4129.41][566][2021.57][93846.00][95177]width=4) (actual time=167.979..11304.413 rows=100000
loops=1)<br/>         Recheck Cond: ((a > 15900000) AND (b > 15900000))<br />         ->  Bitmap Index Scan on
ab_100([0.00][1997.77][566][0.00][1997.77][95177]width=0) (actual time=120.127..120.127 rows=100000 loops=1)<br
/>              Index Cond: ((a > 15900000) AND (b > 15900000))<br /> Total runtime: 11426.329 ms<br />(6
rows)<br/><br />Result for mycorr_10:<br />---------------------------<br /><br
/>Aggregate([4608.36][4608.37][1][94197.91][94197.92][1]width=4) (actual time=24393.058..24393.058 rows=1 loops=1)<br
/>  ->  Bitmap Heap Scan on mycorr_10([2249.51][4606.79][629][2272.83][93963.14][93908] width=4) (actual
time=108.219..24374.050rows=10563 loops=1)<br />         Recheck Cond: ((a > 15900000) AND (b > 15900000))<br
/>        ->  Bitmap Index Scan on ab_10([0.00][2249.35][629][0.00][2249.35][93908] width=0) (actual
time=89.432..89.432rows=10563 loops=1)<br />               Index Cond: ((a > 15900000) AND (b > 15900000))<br
/> Totalruntime: 24393.555 ms<br />(6 rows)<br
/>-------------------------------------------------------------------------------------------------------------<br
/>Goodies:<br/>-----------<br />pg_statio_all_tables -><br />        heap_blks_read=9931     (in case of
mycorr_10)<br/>        heap_blks_read=118693  (in case of mycorr_100)<br /><br />I have repeated the test more than 20
timesup till now.<br />I have also made the same test with different table sizes and correlation level and the same
anomalypersists.<br />Question:<br />------------<br />mycorr_100 took 11.4 s to run although it had to fetch 100000
rowfrom the base table.<br />mycorr_10 took 24.4 s to run although it had to fetch 10563 row from the base table.<br
/><br/>Any explanation for that?<br /><br />Thank you for your patience.<br
/>-------------------------------------------------------------------------------------------------------------<br/><br
/></div><br/><hr size="1" />Ahhh...imagining that irresistible "new car" smell?<br /> Check out <a
href="http://us.rd.yahoo.com/evt=48245/*http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE1YW1jcXJ2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDbmV3LWNhcnM-">new
carsat Yahoo! Autos.</a> 

On Thu, 2007-05-03 at 14:33 -0700, jaba the mobzy wrote:
> mycorr_100 took 11.4 s to run although it had to fetch 100000 row from
> the base table.
> mycorr_10 took 24.4 s to run although it had to fetch 10563 row from
> the base table.

This is because the physical distribution of data is different. The
mycorr_10 table has tuples in which a and b are > 15.9M spread all
throughout. mycorr_100 has them all collected together at the end of the
physical file. Less disk seeking.

You can test this by doing a CLUSTER on both tables and run the same
queries again.

Regards,Jeff Davis

Jeff Davis <pgsql@j-davis.com> writes:
> On Thu, 2007-05-03 at 14:33 -0700, jaba the mobzy wrote:
>> mycorr_100 took 11.4 s to run although it had to fetch 100000 row from
>> the base table.
>> mycorr_10 took 24.4 s to run although it had to fetch 10563 row from
>> the base table.

> This is because the physical distribution of data is different. The
> mycorr_10 table has tuples in which a and b are > 15.9M spread all
> throughout. mycorr_100 has them all collected together at the end of the
> physical file. Less disk seeking.

If the OP had generated the data randomly, as claimed, the rows
shouldn't be particularly more clumped in one table than the other.
But I sure agree that it sounds like a nonrandom distribution in the
mycorr_100 table.  FWIW I tried to duplicate the behavior, and could
not, using tables made up like this:

create table src as
select int4(16*1024*1024*random()) as key,
int4(16*1024*1024*random()) as a,
int4(16*1024*1024*random()) as b
from generate_series(1,16*1024*1024);

create table mycorr_10 as
select key, a,
case when random() < 0.1 then a else b end as b
from src;

create table mycorr_100 as
select key, a, a as b
from src;

create index mycorr_10i on mycorr_10(a,b);

create index mycorr_100i on mycorr_100(a,b);

vacuum analyze mycorr_10;

vacuum analyze mycorr_100;
        regards, tom lane