Hi Jie,
Yeah, basically gather as many stats as I can to accurately profile the
overall system performance. I thought it would be appropriate to use a
TPC-H based workload as one measuring stick to use for bitmap indexes.
Mark
Jie Zhang wrote:
Hi Mark,
Thanks for doing the test. I checked out the link you provided below. I am a
little confused about the goal of these tests. Do you plan to test the
overall performance of postgreSQL on handling TPC-H queries?
Thanks,
Jie
On 9/22/06 3:45 PM, "Mark Wong" <[EMAIL PROTECTED]> wrote:
Jie Zhang wrote:
Hi Heikki and all,
I just sent the latest bitmap index patch to the list. I am not sure if
there is any size limit for this mailing list. If you have received my
previous email, please let me know.
Hi Jie,
I know I said I was going to get testing on this months ago but I've
been juggling between 3 systems due to disk failures and other hardware
configuration issues. Anyways, I've take a baseline run of only the
power test using a 1GB database with the patch 09-17 patch against a
snapshot of pgsql from 2006-09-17:
http://dbt.osdl.org/dbt/dbt3testing/results/dev8-007/2/
Do you think the 1GB scale factor will be sufficient for testing as it
will certainly be faster? Do you think testing with just a power test
will be sufficient for now? I really don't have a good reason why I
didn't run a throughput test other than to save time. :) I also wanted
to get your opinion again on which indexes we will want to try first.
Thanks,
Mark
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
