Subject | Re: [IB-Architect] Next ODS change (was: System table change) |
---|---|
Author | Jason Chapman |
Post date | 2000-11-15T07:58:03Z |
Ann,
See my other post RE: Blob storage.
JAC.
See my other post RE: Blob storage.
JAC.
----- Original Message -----
From: Ann Harrison <harrison@...>
To: <IB-Architect@egroups.com>
Sent: 14/11/2000 4:23 PM
Subject: Re: [IB-Architect] Next ODS change (was: System table change)
> At 11:08 AM 11/14/2000 -0500, Dalton Calford wrote:
> >I have a question (please don't shoot me for asking)
> >
> >Why don't we think about a larger page size?
> >16k, 32k, 64k (these are all natural page read/write sizes on hardware
> >RAID devices and some software RAID systems - the larger block sizes are
> >also supported by various file systems)
> >
> >This would allow for larger key sizes per page and perhaps in certain
> >instances, increase performance as the interbase page size would more
> >closely fit the underlying structures.
> >
> >What would be involved in this?
>
> Not much. My concern is that testing showed a performance
> improvement going to 4K from 1 or 2, but almost none going
> from 4 to 8. We may need to improve the efficiency of some
> page search algorithms.
>
> How many of you consider that your applications are I/O bound?
>
> Regards,
>
> Ann
>
>
> To unsubscribe from this group, send an email to:
> IB-Architect-unsubscribe@onelist.com
>
>
>
>