On 3/31/2017 4:30 PM, Jim Gettys wrote: > > > On Fri, Mar 31, 2017 at 6:18 PM, Joe Touch <to...@isi.edu > <mailto:to...@isi.edu>> wrote: > > > > On 3/31/2017 7:47 AM, Jim Gettys wrote: > > One full size packet @ 1mbps == 13 milliseconds. > That sort of blocking is its own problem and the only solution is > to use > smaller packets. > > It can be solved at TCP by pushing the MSS down, but for UDP that's > another good reason for considering UDP options as a TSVWG WG item! > > > Yes, but without better packet scheduling such as fq_codel at the > bottleneck, it won't help. But I'm decently optimistic that that > fq_codel/cake/other things will be deploying widely, since we're > seeing fq_codel appear in commercial products at this date.
No amount of scheduling overcomes a limit on the unit of scheduling - the packet size, which has increasing impact as link bandwidth decreases. Better scheduling is important necessary too, but by itself insufficient. Joe