Re: Long Running Update - My Solution

Harry Mantheakis <harry.mantheakis@riskcontrollimited.com> wrote:

> I am glad to report that the 'salami-slice' approach worked nicely
> - all done in about 2.5 hours.

Glad to hear it!

> The mystery remains, for me: why updating 100,000 records could
> complete in as quickly as 5 seconds, whereas an attempt to update
> a million records was still running after 25 minutes before we
> killed it?

If you use EXPLAIN with both statements (without ANALYZE, since you
probably don't want to trigger an actual *run* of the statement), is
there a completely different plan for the range covering each?  If
so, a severe jump like that might mean that your costing parameters
could use some adjustment, so that it switches from one plan to the
other closer to the actual break-even point.

> One thing remains crystal clear: I love Postgresql :-)

:-)

-Kevin