[Sis-csi] Text for Transport Layer Section

[Lloyd Wood]

Some comments:
At Tuesday 2006-04-25 14:37 -0400, Scott, Keith L. wrote
"UDP packets may be lost or duplicated in the network, and no 
feedback of such events is provided to the sender."

I suggest:

UDP packets may be lost or duplicated in the network, and no feedback 
of such events is provided to the sender by UDP receivers without use 
of higher-layer protocols.

and later, in the TCP section:

After TCP packets have been lost or duplicated in the network, TCP's 
internal acknowledgement mechanism enables correct handling to 
preserve the bytestream.

(The implication from the original text is that TCP packets are never 
lost or duplicated...)

also:

in stressed environments characterized EITHER by large 
bandwidth*delay products, high bit error rates, OR significant 
asymmetries in data rate

...as any one is sufficient to cause TCP grief.

"IP assumes that network paths run uninterrupted from sender to receiver."

I'd argue that's a function of most routing protocols using IP. If 
you've got static routes out of and deep queues on an interface, you 
could just store the IP packets until the link comes up...

"The common approach to providing enhanced services such as reliable 
multicast or communication without an end-to-end path is to create a 
new layer of protocol on top of either TCP or UDP.  "

...even though you've already stated that TCP peers and expects an 
end-to-end path, and TCP is not suited for multicast? http is a 
better example of an enhanced service built on TCP.

" CFDP can run over TCP"

has anyone ever used CFDP over TCP?

L.




<http://www.ee.surrey.ac.uk/Personal/L.Wood><L.Wood at surrey.ac.uk>