On Thu, Mar 13, 2014 at 11:24 AM, Timothy Baldridge <tbaldri...@gmail.com>wrote:
> I talked to Martin after a CodeMesh, and had a wonderful discussion with > him about performance from his side of the issue. From "his side" I mean > super high performance. > [...] Hi Tim, Thanks for explaining the context of Martin's work - I did not know that it was so advanced. I did some research back in good times on http://en.wikipedia.org/wiki/Connection_Machine (via telnet 56kb/s telnet connection from east Europe to US - that was super cool back then) and can only atest to what he says in the presentation. You can achieve amazing levels of performance provided that patterns of data flow and control in software match exactly parallel nature of given architecture. However his point is somewhat more serious. He directly "attacks" a premise that using immutable data is inherently mutli-core friendly which comes from the deduction that writing reactive software reduces the use locks and guards hence enables more flow and less waiting. Now his point is that GC acts a super GIL which effectively kills all the hard work done on the language and application design level. Now, I wish Marin eats his own dog food and as pointed out numerous times in the presentations, he backs up his points with a real experiments and data. At least it was not apparent wether his conclusions were purely theoretical or grounded in some experience. I am in the process of transitioning from Scala to Clojure ecosystem (just finished SICP videos and have some hard 4clojure problems behind me, but a lot of to learn) so not yet fluent in all aspects of the language but I think at some point I will try to write some simulations of STM performace following some of Martin's intuitions. I think that it will be very cool. Best, Andy -- 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 --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
