Historical Disk bandwidth utilization

Early this morning our sar reports show that WIO on the system was over 50% for about an hour. We also had some users complain about response time problems during this time. Is there a way I can go back and check what disks were busy during this time (something like topas but for historical data)?

the nmon can be enabled to collect data and kep for future reference. you can consider this option, however i am not sure abt historical data... check errpt for any problemaic disks.