Lars Eggert wrote:
On 2008-2-18, at 17:22, ext Eliot Lear wrote:Yes, this is why I did NERD. But let's please make a distinction between delay and drop. I think they can and should be viewed separately.From a transport perspective, it matters whether delay can cause packet reordering to happen or not.
Right. But if the delay is sparse (every few minutes) and we can limit discussion to the "middle" of a communication, this begins to sound like a tractable problem, if it even is noticeable. There are still odd ducks out there, and right now DNS is the biggest thing I see quacking.
If not, the only effect will be, er, more delay. If yes, to most transports the reordering will look like a loss and cause a congestion control response. Which will decrease performance more than delay without reordering would.
And now we have to ask ourselves under what conditions such a thing would happen. If it happens during a routing transition, perhaps it is a reasonable cost. If it just Happens, perhaps not so much.
Eliot -- to unsubscribe send a message to rrg-request@psg.com with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg