RE: fragmentation factor

David, You misundertand my intention. My aim is to provide a comparison figure between different sites so that they have a yardstick to measure by. But my original suggestion does not take into account the likelihood of access. That should be borne in mind. Does it matter if a table is fragmented if it is rarely accessed? Regards Malcolm -----Original Message----- From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org] On Behalf Of david@smooth1.co.uk Sent: 08 October 2005 23:19 To: informix-list@iiug.org Subject: Re: fragmentation factor just get the tablename and number of extents per table...sort by number of extents decending... a simple report like that allows you to see problems... sending to informix-list

"malcolm weallans" <malcolm.iiug@btopenworld.com> wrote in message news:1128939349.1e2a7098a8d96562b9303493f7e99638@teranews... > > David, > You misundertand my intention. My aim is to provide a comparison figure > between different sites so that they have a yardstick to measure by. But > my > original suggestion does not take into account the likelihood of access. > That should be borne in mind. Does it matter if a table is fragmented if > it > is rarely accessed? > > Regards > > Malcolm Does it matter if a table is fragmented anyway post v5? There was a thread on this a month or so back but I don't recall seeing any definitve reasons that it does.

Well it is hard to say ...depends upon disk seek times that are impossible to measure with sans's etc. I'd say it is safer to reduce fragmentation. The problem with fragementation is that it shows people have not properly sized their database. I've found that fragmentation and running out of space tend to go together. I'm not sure any tools replaces analysing each table and sizing it properly.