Re: [PERFORM] COPY insert performance

2005-07-26 Thread Chris Isaacson
lto:[EMAIL PROTECTED] Sent: Tuesday, July 26, 2005 7:12 AM To: Chris Isaacson Cc: pgsql-performance@postgresql.org Subject: Re: [PERFORM] COPY insert performance Hi Chris, Have you considered breaking the data into multiple chunks and COPYing each concurrently? Also, have you ensured that

Re: [PERFORM] COPY insert performance

2005-07-26 Thread Chris Isaacson
shared_buffers would take care of this. I'll increase work_mem to 512MB and rerun my test. I have 1G of RAM, which is less than we'll be running in production (likely 2G). -Chris -Original Message- From: John A Meinel [mailto:[EMAIL PROTECTED] Sent: Monday, July 25, 2005 6

Re: [PERFORM] COPY insert performance

2005-07-26 Thread Chris Isaacson
ork_mem to 512MB and rerun my test. I have 1G of RAM, which is less than we'll be running in production (likely 2G). -Original Message- From: John A Meinel [mailto:[EMAIL PROTECTED] Sent: Monday, July 25, 2005 6:09 PM To: Chris Isaacson; Postgresql Performance Subject: Re: [PERFORM

[PERFORM] COPY insert performance

2005-07-25 Thread Chris Isaacson
Title: Message I need COPY via libpqxx to insert millions of rows into two tables.  One table has roughly have as many rows and requires half the storage.  In production, the largest table will grow by ~30M rows/day.  To test the COPY performance I split my transactions into 10,000 rows.  I