
Does anybody ever use it? I thought it'd speed up my program... instead it turned a 5 minute job into a 100 hour one :s Sofie -- Sofie Van Landeghem PhD Student VIB Department of Plant Systems Biology, Ghent University Bioinformatics and Evolutionary Genomics Technologiepark 927, 9052 Gent, BELGIUM Tel: +32 (0)9 331 36 95 fax:+32 (0)9 3313809 Website: http://bioinformatics.psb.ugent.be

gc() is only useful if you need to reclaim huge objects before you allocate a new one. It can be a very expensive operation with many small objects, only works nicely with few big objects. Other than that, my recommendation is to never ever use it. -T Sofie Van Landeghem wrote:
Does anybody ever use it? I thought it'd speed up my program... instead it turned a 5 minute job into a 100 hour one :s
Sofie

I normally don't use it either, but I ran out of memory this morning because of lazy garbage collection. After that I inserted it in all the new methods to avoid it would happen again... fortunately I didn't actually wait 100 hours before realising something was odd :D Cheers, Sofie Thomas Abeel wrote:
gc() is only useful if you need to reclaim huge objects before you allocate a new one. It can be a very expensive operation with many small objects, only works nicely with few big objects.
Other than that, my recommendation is to never ever use it.
-T
Sofie Van Landeghem wrote:
Does anybody ever use it? I thought it'd speed up my program... instead it turned a 5 minute job into a 100 hour one :s
Sofie
_______________________________________________ Binari Implicitly Neglects All Recursive Iterations https://maillist.psb.ugent.be/mailman/listinfo/binari
-- Sofie Van Landeghem PhD Student VIB Department of Plant Systems Biology, Ghent University Bioinformatics and Evolutionary Genomics Technologiepark 927, 9052 Gent, BELGIUM Tel: +32 (0)9 331 36 95 fax:+32 (0)9 3313809 Website: http://bioinformatics.psb.ugent.be

google gc tuning (on the sun website) if you want to do something about your garbage. I've looked into it quite some time ago, all I remember is GCTimeRatio, something aobout gcpausemilliseconds, -UseParallelGC and lot's of fun size arguments like maxpermsize, permsize, xss, xms, xmx,..., heapfreeratio and something about incrementalmode the lesson i learned was that every project was different. But it did speed up my big-assed eclipse in my equally big-assed windows vista setup... (sharing 2GB amongst vista and eclipse, it's a nightmare, i tell you) On Fri, 2010-06-11 at 14:39 +0200, Sofie Van Landeghem wrote:
I normally don't use it either, but I ran out of memory this morning because of lazy garbage collection. After that I inserted it in all the new methods to avoid it would happen again... fortunately I didn't actually wait 100 hours before realising something was odd :D
Cheers, Sofie
Thomas Abeel wrote:
gc() is only useful if you need to reclaim huge objects before you allocate a new one. It can be a very expensive operation with many small objects, only works nicely with few big objects.
Other than that, my recommendation is to never ever use it.
-T
Sofie Van Landeghem wrote:
Does anybody ever use it? I thought it'd speed up my program... instead it turned a 5 minute job into a 100 hour one :s
Sofie
_______________________________________________ Binari Implicitly Neglects All Recursive Iterations https://maillist.psb.ugent.be/mailman/listinfo/binari
-- Sofie Van Landeghem PhD Student VIB Department of Plant Systems Biology, Ghent University Bioinformatics and Evolutionary Genomics Technologiepark 927, 9052 Gent, BELGIUM Tel: +32 (0)9 331 36 95 fax:+32 (0)9 3313809 Website: http://bioinformatics.psb.ugent.be _______________________________________________ Binari Implicitly Neglects All Recursive Iterations https://maillist.psb.ugent.be/mailman/listinfo/binari
-- ================================================================== Marijn Vandevoorde Tel:+32 (0)9 331 36 95 fax:+32 (0)9 3313809 VIB Department of Plant Systems Biology, Ghent University Technologiepark 927, 9052 Gent, BELGIUM marijn.vandevoorde@psb.ugent.be http://bioinformatics.psb.ugent.be ================================================================== "Stop! Hammer Time!" --M.C. Hammer
participants (3)
-
Marijn Vandevoorde
-
Sofie Van Landeghem
-
Thomas Abeel