Re: Index Scans become Seq Scans after VACUUM ANALYSE

Somethings wrong with the random numbers from the sun... re-run them,
that first sample is insane.... Caching looks like it's affecctign your
results alot...

Mark Pritchard wrote:

>I threw together the attached program (compiles fine with gcc 2.95.2 on
>Solaris 2.6 and egcs-2.91.66 on RedHat Linux 6.2) and ran it a few
>times. Data is below. Usual disclaimers about hastily written code etc
>:)
>
>Machine = ghoul (generic intel, 384mb ram, dual p3-800, ide disk running
>dma)
>
>Sequential
>Bytes Read Time Bytes / Sec
>536870912 27.14 19783933.74
>536870912 27.14 19783990.60
>536870912 27.11 19801872.14
>536870912 26.92 19942928.41
>536870912 27.31 19657408.43
> 19794026.66 (avg)
>
>Random
>Bytes Read Time Bytes / Sec
>1073741824 519.57 2066589.21
>1073741824 517.78 2073751.44
>1073741824 516.92 2077193.23
>1073741824 513.18 2092333.29
>1073741824 510.68 2102579.88
> 2082489.41 (avg)
>
>Machine = jedi (Sun E420, 3gb ram, dual 400s, test on single scsi disk)
>
>Sequential
>Bytes Read Time Bytes / Sec
>2097152000 65.19 32167675.28
>2097152000 65.22 32154114.65
>2097152000 65.16 32182561.99
>2097152000 65.12 32206105.12
>2097152000 64.67 32429463.26
> 32227984.06 (avg)
>
>Random
>Bytes Read Time Bytes / Sec
>4194304000 1522.22 2755394.79
>4194304000 278.18 15077622.05
>4194304000 91.43 45874730.07
>4194304000 61.43 68273795.19
>4194304000 54.55 76890231.51
> 41774354.72
>
>If I interpret Tom's "divide" instruction correctly, is that a factor of
>10 on the linux box?
>