> There seems to be a problem in the way memory is used when running Aria
> using CCPN data. I have run several aria jobs using my analysis project
> and writing the data back into analysis at the end. It has now got to
> the stage where starting Aria takes 24 hours on my brand new 2GB of
> memory dual processor machine, although this is only 7-8 minutes of
> elapses CPU on the python job. My analysis project directory (ie not
> including the spectra directory) is 637MB compared to about 250MB for
> the project before. However when I start an aria job python fills up
> the 2GB and much of the swap space and crawls. Once the CNS jobs start
> some of the memory is freed up and performance is OK, but even then
> python has large amounts of memory.
Hi Nick,
I presume that the large increase in the CCPN project size is expected
because you are saving lots of data from the ARIA run (structures,
restraint lits etc). If so, this ought not effect things too much in the
future because a CCPN project only loads what it needs, and you'll
probably never want look at everything.
I guess what is happening is that the CCPN project memory is not being
unallocated each time you get output from an ARIA iteration. Although,
I'll have to investigate (probably with the assistance of Wolfgang) to be
sure. It might be that ARIA reopens the CCPN project each iteration and
the CCPN memory from the previous iteration is still hanging around. If
this is not the case then it could be that all the output restraint lists
and structures from every iteration are hanging about because they are not
unloaded.
You could perhaps try running ARIA without writing any CCPN data except
during the last iteration. If this works quickly, without eating into your
swap space, then I think that would clarify things.
In any case we'll have a play at this end.
T.
-------------------------------------------------------------------------------
Dr Tim Stevens Email: [log in to unmask]
Department of Biochemistry [log in to unmask]
University of Cambridge Phone: +44 1223 766018 (office)
80 Tennis Court Road +44 7816 338275 (mobile)
Old Addenbrooke's Site +44 1223 364613 (home)
Cambridge CB2 1GA WWWeb: http://www.bio.cam.ac.uk/~tjs23
United Kingdom http://www.pantonia.co.uk
-------------------------------------------------------------------------------
------ +NH3CH(CH(CH3)OH)C(O)NHCH(CH(CH3)CH2CH3)C(O)NHCH(CH2CH2SCH3)CO2- -------
-------------------------------------------------------------------------------
|