Re: DBA_HIST_SYSMETRIC_HISTORY and metric I/O Megabytes per Second
From: David Mann <dmann99_at_gmail.com>
Date: Thu, 12 May 2011 22:33:25 -0400
Message-ID: <BANLkTimOeZN1CC40a5tznhE4zcPyOGSOJA_at_mail.gmail.com>
>Hi all
>Anyone know if the metric I/O Megabytes per Second is valid in
>DBA_HIST_SYSMETRIC_HISTORY at version 11gR1?
>I am checking some I/O throguhput in a pretty loaded system (around
>1200MB/sec) and this metric shows some really low value like 6MB/sec!
Date: Thu, 12 May 2011 22:33:25 -0400
Message-ID: <BANLkTimOeZN1CC40a5tznhE4zcPyOGSOJA_at_mail.gmail.com>
>Hi all
>Anyone know if the metric I/O Megabytes per Second is valid in
>DBA_HIST_SYSMETRIC_HISTORY at version 11gR1?
>I am checking some I/O throguhput in a pretty loaded system (around
>1200MB/sec) and this metric shows some really low value like 6MB/sec!
How is the storage with your datafiles attached? I was looking for similar information in OEM to match up with iostat numbers I was observing on some NFS mounts... but at least in this case OEM only 'knew' about locally attached drives, not the NFS mounts where the real action was happening. So I observed something similar, busy database but OEM was only telling me about the light activity going on locally (Oracle binaries and OS activity). Maybe we are seeing something similar?
-Dave
-- Dave Mann www.brainio.us www.ba6.us - Database Stuff - http://www.ba6.us/rss.xml -- http://www.freelists.org/webpage/oracle-lReceived on Thu May 12 2011 - 21:33:25 CDT