On Sun, May 18, 2014 at 10:09 AM, Ron Johnson <ron.l.johnson@cox.net> wrote:
> 10% packet loss is acceptable. :(
That's total nonsense. 0.10 packet loss between hosts at reasonable
latencies easily degrades TCP/IP performance by 80% or more. Should
ask.... "So then it should be acceptable if I only pay the bill 20%
of the time?"
For example, at an 100ms round trip delay (ping time), for 10% packet
loss; the maximum theoretical possible throughput you can achieve is
scaled down by the loss --- the maximum sustained TCP throughput is
limited to approximately 0.3 Megabits at 10%/100ms.
> --
> "Mathematics deals exclusively with the relations of concepts to each
> other without consideration of their relation to experience."
> Albert Einstein
> ___________________
> Nolug mailing list
> nolug@nolug.org
-- -JH ___________________ Nolug mailing list nolug@nolug.orgReceived on 05/18/14
This archive was generated by hypermail 2.2.0 : 05/18/14 EDT