What sysmaster table?

Folk: Does anyone know if the information on the onstat -g iof report for the IO service times for each device/chunk/file are available in sysmaster at all? I can't find that information, but maybe I'm just being dense or senior. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. --001636283b14aff6a204925ac67b

Art Kagel Wrote: ============================================================================== Folk: Does anyone know if the information on the onstat -g iof report for the IO service times for each device/chunk/file are available in sysmaster at all? I can't find that information, but maybe I'm just being dense or senior. Art ============================================================================== Response: This query gets very close. See how it measures up on your systems. Population of readtime / writetime may depend on version. select ((reads + writes) * 999600000/(readtime+writetime)) from syschktab where readtime+writetime <> 0 Dave Griffen

That's pages/usec which isn't what I want, I want service time in fractional seconds, but you are correct, all of the data is there and I'm just being dense! Here's my query: select dbsnum, chknum, fname, reads, readtime/1000000 read_time_secs, case when reads > 0 then (readtime / 1000000 / reads) else 0 end as read_service_secs, writes, writetime/1000000 write_time_secs, case when writes > 0 then (writetime / 1000000 / writes) else 0 end as write_service_secs from syschktab order by 1, 2; These numbers agree with the onstat -g iof output - YEA! Thanks Dave for putting me on the right track! Note that the times include all time the engine spent waiting for disk for that chunk, so if you have initialized the instance or added a chunk since the stats were last zero'd, the wait time to verify/validate the chunk (and possibly allocate the disk space if the chunk is in filesystem space) is included so the service times for those new chunks will be huge. But they still agree with onstat. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Tue, Oct 12, 2010 at 1:06 PM, DAVE GRIFFEN <dgriffen@finishline.com>wrote: > Art Kagel Wrote: > > ============================================================================== > Folk: > > Does anyone know if the information on the onstat -g iof report for the IO > service times for each device/chunk/file are available in sysmaster at all? > I can't find that information, but maybe I'm just being dense or senior. > > Art > > ============================================================================== > > Response: > This query gets very close. See how it measures up on your systems. > Population > of readtime / writetime may depend on version. > > select ((reads + writes) * 999600000/(readtime+writetime)) > from syschktab > where readtime+writetime <> 0 > > Dave Griffen > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --0014853c975ed4616104926f9426