Both after your message and doing some more thinking, I'm sure that I can just
use all of the data. A general average will be good enough for me after I
thought this through a bit more.
I'm basically thinking about the possibilities of a backup retention scheme
where I keep all of my incrementals for a long period, but not the fulls. I
need to determine how much media I will go through just for those incrementals.
As you said, if I missed a incremental, the next day's will be larger, but can
be averaged across both days.
I also made a rule in an Excel spreadsheet to only include days when there was
a successfull backup the previous day, and then average across the count of
those occurances (instead of across the total # of days during the period).
This will give me a different perspective to compare to also. I'll obviously
sort by incremental size and see how it looks if I trim off the extremes.
Thanks,
Jason
-----Original Message-----
From: Tim Mooney [mailto:mooney AT DOGBERT.CC.NDSU.NODAK DOT EDU]
Sent: Thursday, September 26, 2002 3:52 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] incremental backup size reporting
In regard to: [Networker] incremental backup size reporting, Jason Heller...:
>I basically want to determine what my average daily incremental backup size
>is, based on past backups. The complex part about this is, I want to be
>sure it is xB/day - so the report would need to filter out incrementals
>that were not preceded by a successful backup the day before, otherwise,
>the size would be inaccurate.
You want to drop them completely (i.e. not count them at all, which would
mean your monthly totals would underreport how much you backed up), or you
want to somehow account for the fact that "incr for day NN/MM/YYYY includes
changes from NN/MM-1/YYYY too"?
> I'm sure there are other variables that
>would need to be addressed too, like restricting the report to specific
>savesets/clients as to not return inaccurate spikes when new
>savesets/clients were added.
>
>Does anyone have any recommendations for gathering such data? I'm going to
>look into both manual efforts and 3rd party tools/scripts (Bocada, etc.),
>but thought maybe someone here has done this before.
I've done something similar, but not to the point where I treat
"accumulated incrs" specially -- I just assume that if three days were
missed for a particular client then the amount of data that was backed up
during the next incremental should be averaged over the number of days
missed. I *don't* currently have my backup reporting scripts reporting
this average -- they just show `0' MB for one day, and a larger than
average number for the next day, and then I eyeball average it.
Tim
--
Tim Mooney mooney AT dogbert.cc.ndsu.NoDak DOT edu
Information Technology Services (701) 231-1076 (Voice)
Room 242-J6, IACC Building (701) 231-8541 (Fax)
North Dakota State University, Fargo, ND 58105-5164
--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
InterScan_Disclaimer.txt
Description: Text document
|
