|
Oh I see, you were using your group commit type
scheme,
so even with only a few commits per second, you'll
see large
numbers of entries per second.
----- Original Message -----
Sent: Friday, March 04, 2005 11:20
AM
Subject: Re: back-bdb IDL
limitations
The disk used in the
experiment is the Seagate ST336753LC (15K RPM). Write caching was
disabled in the experiment.
-
Jong-Hyuk
------------------------
Jong Hyuk Choi
IBM Thomas J. Watson
Research Center - Enterprise Linux Group
P. O. Box 218, Yorktown Heights,
NY 10598
email: jongchoi@us.ibm.com
(phone)
914-945-3979 (fax) 914-945-4425 TL:
862-3979
From experiments, I didn't see a significant
reduction in the population time by the removal of the transactions. The
experimental data shows instead that it is the number of database operations
which has more effect on the performance of the directory
population.
Sounds like
you tested on a disk with write-back-caching ?
Otherwise you should be seeing only a few tens of
commits per second due to physical disk rotation
speed.
|