[Biopython-dev] Benchmarking PDBParser

[João Rodrigues]

Hey Peter,

I've only had the chance to test this today. The parsing seems to be working
just fine and the RAM consumption is stable at < 100 MB. I'll see the
results tomorrow.

Cheers,

João [...] Rodrigues
http://nmr.chem.uu.nl/~joao



On Fri, May 6, 2011 at 10:29 AM, Peter Cock <p.j.a.cock at googlemail.com>wrote:

> On Fri, May 6, 2011 at 9:24 AM, João Rodrigues <anaryin at gmail.com> wrote:
> >> Memory bloat is bad - it sounds like a garbage collection problem.
> >> Are you recreating the parser object each time?
> >
> > No. I'm just calling get_structure at each step of the for loop. It's a
> bit
> > irregular also, sometimes it drops from 1GB to 300MB, stays stable for a
> > while and then spikes again. My guess is that all the data structures
> > holding the parser structures consume quite a lot and probably there's no
> > decent GC to clear the previous structure in time, so it accumulates.
> >
>
> You could do an explicit clear once per PDB file to test this hypothesis:
>
> import gc
> gc.collect()
>
> Peter
>