On Apr 3, 2012, at 8:21 PM, billywhizz wrote: > thanks for the feedback ben. > > as for process isolation. i agree it's a better way of doing things for most > scenarios but if you want to build a low latency system there is a lot of > overhead in having to serialise objects and pass them across a process > boundary. at some stage, it would be nice to be able to build a system in > node.js that used many cores on one machine and is able to efficiently share > objects/data between them without a lot of locking and contention. as far as > i can see the only way of being able to do this REALLY fast is to use threads > and manage contention using volatile variables or something like that. i have > been doing a lot of reading around the LMAX disruptor pattern which is able > to get orders of magnitude better performance than a traditional process or > thread/mutex based solution. i'd love to see node.js/v8 support for something > like this in the future, although i can see that it would be very difficult > to achieve. > > more info on distruptor here: > > http://code.google.com/p/disruptor/
Of course, in the meantime, you can use <https://github.com/xk/node-threads-a-gogo> :-) It runs JavaScript in threads/pools of threads: - almost twice as fast as node's main thread - with several times the throughput of a cluster of nodes communicating via IPC - needs just a fraction of the memory of a cluster - will be even faster in the near future: it will soon pass data by reference across threads. -- Jorge. -- Job Board: http://jobs.nodejs.org/ Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines You received this message because you are subscribed to the Google Groups "nodejs" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/nodejs?hl=en?hl=en
