Re: Multiple Indexing, performance impact - Mailing list pgsql-general

From Daniel Åkerud
Subject Re: Multiple Indexing, performance impact
Date
Msg-id 003a01c0fb64$32154510$c901a8c0@automatic100
Whole thread Raw
In response to Re: Multiple Indexing, performance impact  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-general
Tried with 2048 also, it complete took away the strange steep after 7:

D is now 2048

1.    A: 36    B: 32    C: 35    D: 31
2.    A: 69    B: 53    C: 38    D: 38
3.    A: 97    B: 79    C: 40    D: 40
4.    A: 131    B: 98    C: 48    D: 43
5.    A: 163    B: 124    C: 52    D: 49
6.    A: 210    B: 146    C: 66    D: 50
7.    A: 319    B: 233    C: 149    D: 58
8.    A: 572    B: 438    C: 268    D: 65
9.    A: 831    B: 655    C: 437    D: 76
10.    A: 1219    B: 896    C: 583    D: 79

What is the program called that flushes the buffers every 30 seconds on a
linux 2.2.x system?

Daniel Åkerud

> > Holy ultra-violet-active macaronies :)
> >
> > First I changed it to 256, then I changed it to 1024.
> >
> > -B 128 is A
> > -B 256 is B
> > -B 1024 is C
> >
> > New multiple-index performance data):
> >
> > 1.    A: 36    B: 32    C: 35
> > 2.    A: 69    B: 53    C: 38
> > 3.    A: 97    B: 79    C: 40
> > 4.    A: 131    B: 98    C: 48
> > 5.    A: 163    B: 124    C: 52
> > 6.    A: 210    B: 146    C: 66
> > 7.    A: 319    B: 233    C: 149
> > 8.    A: 572    B: 438    C: 268
> > 9.    A: 831    B: 655    C:
> > 10.    A: 1219    B: 896    C:
> >
> > The last test hasn't finished yet, but THANKS! I know the reson now, at
> > least... i'll try
> > 2048 also.
>
> Strange that even at 1024 performance still drops off at 7.  Seems it
> may be more than buffer thrashing.
>
>
> > -B equals --brutal-performance ? ;)
>
> See my performance article on techdocs.postgresql.org.
>
> --
>   Bruce Momjian                        |  http://candle.pha.pa.us
>   pgman@candle.pha.pa.us               |  (610) 853-3000
>   +  If your life is a hard drive,     |  830 Blythe Avenue
>   +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026
>


pgsql-general by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Multiple Indexing, performance impact
Next
From: Tom Lane
Date:
Subject: Re: Multiple Indexing, performance impact