> On Tue, Mar 15, 2011 at 9:08 AM, Robert Brewer <fumanchu(at)aminus(dot)org>
> > I'm working on a hypercube implementation in Postgres using
> > and need to insert 80,000 rows in one go from Python. Doing so with
> > INSERT, even multiple statements in one call, is pretty slow. I've
> > investigating if using COPY is faster.
> When you say "multiple statements", do you mean
> INSERT INTO foo (coords) VALUES
> (cube(ARRAY[1, 'nan', 3]::float)),
> (cube(ARRAY[2, 'nan', 4]::float)),
> (cube(ARRAY[3, 'nan', 5]::float)),
> (cube(ARRAY[4, 'nan', 6]::float));
> I was going to suggest trying that method, but if you already have
> then please ignore me!
Yes, I'm using the above now. Looking for something faster. I'll probably settle on using -Inf for the short term so I can use COPY.
In response to
pgsql-bugs by date
|Next:||From: Tambet Matiisen||Date: 2011-03-16 19:08:07|
|Subject: Re: BUG #5929: ERROR: found toasted toast chunk for toast
value 260340218 in pg_toast_260339342|
|Previous:||From: bricklen||Date: 2011-03-16 16:18:02|
|Subject: Re: SELECT '(1, nan, 3)'::cube;|