Re: [HACKERS] strange behavior of UPDATE

Bruce Momjian wrote:
>
> > > Not really. It required changing function calls all over, and probably
> > > relies on 6.5 changes too. It is in the 6.5 beta right now. Are you
> > > using that?
> >
> >
> > I did all the tests with the current snapshot of May 18th.
> >
>
> Fix was installed yesterday.
>

sorry, but the problem still persists. I used

-rw-r--r-- 1 1005 root 5954900 May 24 03:03 postgresql.snapshot.tar.gz

I started the query

update bench set k500k = k500k + 1 where k100 = 30;

and I killed it after half an hour of havy disk activity.
The same query on the same machine with the same setup,
but using sybase-ase-11.0.3.3-1 takes less than 1 minute.

The table bench looks like the following:

Table = bench
+----------------------------------+----------------------------------+-------+
| Field | Type | Length|
+----------------------------------+----------------------------------+-------+
| kseq | int4 not null | 4 |
| k500k | int4 not null | 4 |
| k250k | int4 not null | 4 |
| k100k | int4 not null | 4 |
| k40k | int4 not null | 4 |
| k10k | int4 not null | 4 |
| k1k | int4 not null | 4 |
| k100 | int4 not null | 4 |
| k25 | int4 not null | 4 |
| k10 | int4 not null | 4 |
| k5 | int4 not null | 4 |
| k4 | int4 not null | 4 |
| k2 | int4 not null | 4 |
| s1 | char() not null | 8 |
| s2 | char() not null | 20 |
| s3 | char() not null | 20 |
| s4 | char() not null | 20 |
| s5 | char() not null | 20 |
| s6 | char() not null | 20 |
| s7 | char() not null | 20 |
| s8 | char() not null | 20 |
+----------------------------------+----------------------------------+-------+

The table is filled with 1.000.000 rows of random data
and on every field an index is created.

Edmund

--
Edmund Mergl mailto:E(dot)Mergl(at)bawue(dot)de
Im Haldenhau 9 http://www.bawue.de/~mergl
70565 Stuttgart fon: +49 711 747503
Germany