Re: COPY ... FROM and index usage

Hi,

Anyone have comparisons/benchmarks to give some
idea of the potential performance gains?

Say compared to doing the stuff here:
http://www.postgresql.org/docs/8.2/static/populate.html

Regards,
Link.

At 09:35 AM 11/5/2007, Toru SHIMOGAKI wrote:
>Dimitri, thank you for your quoting. I'm a pg_bulkload author.
>
>pg_bulkload is optimized especially for appending data to table with indexes.
>If you use it, you don't need to drop index
>before loading data. But you have to consider
>conditions carefully as Dimitri said below. See also pg_bulkload README:
>
>http://pgfoundry.org/docman/view.php/1000261/473/README.pg_bulkload-2.2.0.txt
>
>Best regards,
>
>Dimitri Fontaine wrote:
>>Hi,
>>Le Sunday 04 November 2007 11:22:19 Reg Me Please, vous avez écrit :
>>>That is, should I drop all indexes during a "COPY ... FROM" in order to
>>>gain the maximum speed to load data?
>>When looking for a way to speed up data
>>loading, you may want to consider pgbulkload, a
>>project which optimizes index creation while
>>loading data, and bypass constraints: it's
>>useful when you know you trust input.
>>As I've never used it myself, I can only provides following links:
>> http://pgfoundry.org/projects/pgbulkload
>> http://pgfoundry.org/docman/view.php/1000261/456/20060709_pg_bulkload.pdf
>>Regards,
>
>--
>Toru SHIMOGAKI<shimogaki(dot)toru(at)oss(dot)ntt(dot)co(dot)jp>
>NTT Open Source Software Center
>
>
>---------------------------(end of broadcast)---------------------------
>TIP 6: explain analyze is your friend