On Tue, Apr 23, 2013 at 4:15 AM, Mikeal Rogers <[email protected]> wrote: > Is there a reason not to just have the underlying libuv *always* writev when > it has more than one pending buffer to write?
It could but it doesn't (at least, not always.) On most operating systems, it's often faster to call write() two or three times in a row than it is to call writev() once (for small payloads.) There is an open issue[1] and some in-progress work[2] but it only has a significant impact on big writes, on the order of 100 kB or more. [1] https://github.com/joyent/libuv/issues/742 [2] https://github.com/bnoordhuis/libuv/compare/issue742 > I'm wondering whey we can't just optimize this behind the scenes, is there a > reason we need to map each stream write a write syscall? node.js and libuv optimize for latency first, throughput second (because you can always buffer higher up in the application.) IOW, it tries to send out pending data as quickly as possible. (Caveat emptor: I'm not 100% sure if that statement is always correct when streams2 are involved.) Because node.js doesn't know when new data will be sent, it doesn't know when it's sound to buffer. That's why the choice to buffer has to be an explicit one, made by the programmer. -- -- 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 --- You received this message because you are subscribed to the Google Groups "nodejs" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
