Re: Expiration Performance Statistics
2000-01-20 10:41:42
Hi Frederic,
With RS/6000 H70 2GB memory; DB on SSA D40 9GB drives spread over 2
RAID sets (6+HS in each); recovery log (mirrored) on SSA 9GB drives
(different adapters) and with AIX 4.3.2/ADSM 4.1.2.42 : we are seeing
200-240 examines per second. We generally examine 8-9 million files
per week and delete 2-3 million.
We do a single run on a (quietish) Saturday but it isn't much worse on
weekdays, with a bit of other activity. DB is just over 60GB; cache
hit rate > 99%.
It was a lot worse before 3.1.2.42 as low as 20/sec (due to hangs and
generally poorer performance). I would like to see it better : this
is one of the limiting factors to scaling up, and will be (I think)
even more important in the future with SANs etc and the data bypassing
the ADSM server. The ADSM server will then become more of a DB engine.
Sheelagh
------------------------------------------------------------------------
Sheelagh Treweek Email: sheelagh.treweek AT oucs.ox.ac DOT uk
Sheelagh Treweek Email: sheelagh.treweek AT oucs.ox.ac DOT uk
Oxford University Computing Services Tel: +44 (0)1865 273205
13 Banbury Road, Oxford, OX2 6NN, UK Fax: +44 (0)1865 273275
------------------------------------------------------------------------
=========================================================================
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- Expiration Performance Statistics, Medico, Frederic
- Re: Expiration Performance Statistics, Richard Cowen
- Re: Expiration Performance Statistics, Richard Sims
- Re: Expiration Performance Statistics, Medico, Frederic
- Re: Expiration Performance Statistics, Jordan, Chris (ELS)
- Re: Expiration Performance Statistics, Tab Trepagnier
- Re: Expiration Performance Statistics,
Sheelagh Treweek <=
- Re: Expiration Performance Statistics, John Naylor
- Re: Expiration Performance Statistics, Miles Purdy
- Re: Expiration Performance Statistics, Herve Chibois
- Re: Expiration Performance Statistics, Mauro M. TINELLI
- Re: Expiration Performance Statistics, Cole C Rogers
- Re: Expiration Performance Statistics, Richard Sims [mailto:rbs
- Re: Expiration Performance Statistics, Medico , Frederic [mailto:FMedico
|
|
|