On Aug 4, 2010, at 12:35 PM, David Nolen wrote: > 8gb of ram seems kinda low to me for this kind of microbenchmark, especially > when running 16 or 48 cores. But I'm no expert. As a far-fetched comparison > when I played around with Aleph on 8-core box, doing nothing but hitting the > server as fast as possible ate up 2gb of RAM. Seems like there was less > allocation there than here (I couldn't even serve more than 20k reqs a second > much less 1e6).
This helps too! Bumping the RAM up to 16GB ram produces slightly better performance on the 48 core machine with the arithmetic burn function, and even eliminates the bizarre slowdowns on both the 16- and 48-core machines with the garbage-full burn function. So collectively this answers a lot of my questions, telling me I should use lots of RAM, parallel GC, and avoid garbage even more than usual (although this goes against my habits). I'm still interested in any feedback on my "is there a best way -- idiomatically and with respect to performance..." question, and I would still like to better understand where those anomalies came from and how to avoid them, particularly since a lot of my real code will unavoidably generate garbage. And I haven't tried re-running my tests on my dual core macbook yet, which I think will still be slower than sequential... Thanks so much, -Lee -- Lee Spector, Professor of Computer Science School of Cognitive Science, Hampshire College 893 West Street, Amherst, MA 01002-3359 lspec...@hampshire.edu, http://hampshire.edu/lspector/ Phone: 413-559-5352, Fax: 413-559-5438 Check out Genetic Programming and Evolvable Machines: http://www.springer.com/10710 - http://gpemjournal.blogspot.com/ -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
