Hi, Andy. Thanks for the responses and information.
Just to let you know... what we are storing in the db are the image
attributes - about 40 of them - not the images. So the COPY
is reading an ascii file of the image attributes. It turns out to be
useful to have the image attributes handy - m
On 1 Aug 2009, at 23:24, Janet Jacobsen wrote:
My questions are:
(2) Should I REINDEX these two tables daily after the pipeline
completes? Is this what other people do in practice?
No need, but as people are querying as soon as data is arriving, an
analyse of the table you just copied to shou
Janet Jacobsen wrote:
Thanks for your reply. Responses below to answer your questions,
and one follow-up question on REINDEX.
Janet
Both tables have one or two foreign key constraints.
(2) Should I REINDEX these two tables daily after the pipeline
completes? Is this what other people do in
Thanks for your reply. Responses below, and one follow-up
question about when/how often to use analyze.
Janet
On 02/08/2009 05:09 a.m., Alban Hertroys wrote:
On 1 Aug 2009, at 23:24, Janet Jacobsen wrote:
My questions are:
(2) Should I REINDEX these two tables daily after the pipeline
comple
On 1 Aug 2009, at 23:24, Janet Jacobsen wrote:
Both of the big tables (now up to > 15 M rows each) have
indexes on several of the columns. The indexes were
created using CREATE INDEX CONCURRENTLY...
Both tables have one or two foreign key constraints.
Others have answered some of your questio
Andy Colson writes:
> I have never tried two processes copy'ing into the same table at the same
> time, but I'd bet its possible.
It's no different from concurrent INSERTs ...
regards, tom lane
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To ma
Janet Jacobsen wrote:
Hi. We are running a data processing/analysis pipeline that
writes about 100K records to two tables on a daily basis.
The pipeline runs from about 6:00 a.m. to 10:00 a.m.
Our user base is small - about five people. Each accesses
the database in a different way (generally
Hi. We are running a data processing/analysis pipeline that
writes about 100K records to two tables on a daily basis.
The pipeline runs from about 6:00 a.m. to 10:00 a.m.
Our user base is small - about five people. Each accesses
the database in a different way (generally using some script
- eit
