First, I'd like to mention that sqlobject is one of my all-time
favorite python packages. Job well done!
However, for my latest project, I'm working on a worldwide monitoring
system that tracks a large number of parameters. It takes in about
40,000 parameters / minute (thank god for huge RAID 5 arrays to write
to). I initially implemented this all using SQLObject to interface
to the DB backend for convenience. Once I started to ramp up the
system, I noticed some bad memory usage.
After a couple hours of running, the python process takes up about 2
GB of RAM. I took SQLObject out and put direct calls to the DB in,
and the memory usage is < 50 MB.
Any hints? I tried "del <object>" after I'm through with each of
them. Is there some special connection handling I should do to get
SQLObject to release the memory?
Thanks a bunch!
On Sat, Aug 05, 2006 at 12:06:57AM -0500, Brian Bockelman wrote:
> After a couple hours of running, the python process takes up about 2
> GB of RAM. I took SQLObject out and put direct calls to the DB in,
> and the memory usage is < 50 MB.
From time to time clear the cache:
Oleg Broytmann http://phd.pp.ru/ phd@...
Programmers don't die, they just GOSUB without RETURN.