The benchmarks referred to in the article are a bit dated, but dmalloc still makes the results more comparable (although a gap remains). For the current benchmarks see:
http://sekhon.berkeley.edu/macosx I just had dinner last night with someone in Apple's Mac OS X performance group, and the default malloc for 10.5 will be better optimized for our purposes. The BLAS implementation in the vecLib Framework will also be significantly improved for x86. Cheers, Jas. ======================================= Jasjeet S. Sekhon Associate Professor Travers Department of Political Science Survey Research Center UC Berkeley http://sekhon.berkeley.edu/ V: 510-642-9974 F: 617-507-5524 ======================================= Andrew Beckerman writes: > Dear all - > > Has anyone seen the discussion surrounding this? They test they > refer to uses R on Windows, Linux and OSX. The upshot appears to be > that building R against the Lea malloc on OSX, rather than the > default ,makes for a more comparable result.... > > http://www.serverwatch.com/news/article.php/3609506 > > Cheers > andrew > > > > ------------------------------------------------------------------------ > --------- > Dr. Andrew Beckerman > Department of Animal and Plant Sciences, University of Sheffield, > Alfred Denny Building, Western Bank, Sheffield S10 2TN, UK > ph +44 (0)114 222 0026; fx +44 (0)114 222 0002 > http://www.beckslab.staff.shef.ac.uk/ > ------------------------------------------------------------------------ > ---------- > > _______________________________________________ > R-SIG-Mac mailing list > [email protected] > https://stat.ethz.ch/mailman/listinfo/r-sig-mac _______________________________________________ R-SIG-Mac mailing list [email protected] https://stat.ethz.ch/mailman/listinfo/r-sig-mac
