On 14 September 2010 12:46, Martin Sustrik <[email protected]> wrote:

> On 09/14/2010 06:40 PM, Steven McCoy wrote:
>
>  The timeout is more likely in the range milliseconds to microseconds for
>> rate limiting.  0.1s can be valid for recovery.
>>
>
> Ok, so we have 2 windows. One for "already sent stuff" (recovery buffer)
> another for "stuff to be sent" (rate limiting buffer). Right?
>
> What's the size of the latter one then? Can it be possibly sized so that it
> has rate*0.1s length?
>
>
The latter is one sequence number in size otherwise you can get the
situation that receivers are asking for re-transmission of packets that
haven't been sent yet.  Timer based TX flushing puts transmission time into
the receiver thread for bus transports which will only be adverse for
performance.

-- 
Steve-o
_______________________________________________
zeromq-dev mailing list
[email protected]
http://lists.zeromq.org/mailman/listinfo/zeromq-dev

Reply via email to