Jodi,
I believe DBI defaults to autocommitting queries, which slows it down. If you aren't specifically turning off the autocommit, I'd suggest it, which would make all the inserts go in one transaction which makes it much faster. You'd do something like:
DBI->connect("dbi:Pg:dbname=dbname;host=hostname", 'user', 'password', {AutoCommit => 0})
Thanks,
Peter Darley
I am currently using a Perl data loader that was set up to load data to three particular tables. The majority of this data is entered into one table. The scripts obtain the data by parsing it out of an Excel spreadsheet. My loads seem to take a very long time. Each file only has about 12,000 rows, yet it takes close to 25 minutes to load one file. I have placed some debugging syntax in the code and it seems that the extra time if related to postgres as I had originally thought it may have to do with the parsing of the Excel file.
I have tried turning off FSYNC but there is no difference in load time. I also tried removing the indexes from the table in which most of the data is loaded but no luck. By the way, is there anyway to confirm that the FSYNC option was turned off correctly? a way to view status or something?
Any thoughts on what might be going on? another performance tuning trick that I have not thought of?
Thanks for your help,
Jodi Kanter
_______________________________
Jodi L Kanter
BioInformatics Database Administrator
University of Virginia
(434) 924-2846
jkanter@virginia.edu
