Thanks isaac and laurent for your answers. My system is based on a Intel P4 (32 bit) with 2Go Ram & 2Go Swap, It's a Linux 2.4.18-3 compiled as 32 bit, and Redhat 7.3 R is compiled as 32 bit. I have no other user process runing during the analysis. I have somes questions : - Does it's help if I rebuild R with a 64 bit option without rebuild the kernel and increase the swap ? (I don't think so) - Does it's help if I rebuild R and linux with 64 bit without changing the computer ? (I don't think so ...) - If I need to change the computer (32 -> 64 bits), I'm looking for a "cheaper" configuration with linux, R, (and of course BioConductor, affy, ;-). Do you have somes recomendations ? TIA, L. Buffat -----Message d'origine----- De : isaac.neuhaus@bms.com [mailto:isaac.neuhaus@bms.com] Envoyé : jeudi 27 février 2003 19:18 Ŕ : Laurent Buffat Cc : bioconductor@stat.math.ethz.ch Objet : Re: [BioC] Out of memory during ReadAffy I don't know what system you are running but if you can build 64 bit R that is not a problem. I have run 254 HG-U95Av2 chips whit out a problem. Isaac Laurent Buffat wrote: > Hi, > > I'm trying to analyze 120 arrays (Affy 133A chip) simultaneously, but I get > an allocation error with R during ReadAffy > > My computer has 2 GO of Ram and 2 Go of swap. > > Do you think it is possible? > Has anyone of you already tried (succesfully ;-) ) to analyze at the same > time a lot of chips (more than 100)? > > TIA > > Laurent Buffat. > > ////////////// > > Laurent Buffat MD, PhD > It-Omics > Parc Eurasanté > 885, avenue Eugčne Avinée > 59120 Loos > Tél : 33 (0)3 20 16 40 56 > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor