I've been running dbcc checks against an upgraded (from 11.5) database on
12.5.0.3. I've run update statistics on the database beforehand. The
sp_sysmon shows high I/O during the dbcc checkstorage which I'm sure shouldn
't be so high. Sybase Support has suggested that some large tables could be
fragmented (optdiag is showing this is the case) but I'm wondering if the
high I/O is due to me having the sybase devices in a simple AIX mirrored
logical volume setup rather than a RAID5 scenario. Is it normal for large
tables to become fragmented during an upgrade? Perhaps they were fragmented
beforehand (but with no optdiag I cant tell) but the dbcc checkstorage on
the 11.5 system hasnt been reporting I/O problems.

Any suggestions are welcome.

Regards,
Martin