Sorry to follow up on my on post.  Meant to change a comment before I sent...

On Mon, Sep 5, 2011 at 3:05 PM, Bruce Lowekamp <[email protected]> wrote:
> On Sun, Sep 4, 2011 at 8:50 AM, SCHARF, Michael
> <[email protected]> wrote:
>>
>> Section 5.6: "The Framing Header (FH) is used to frame messages and
>> provide timing when used on a reliable stream-based transport protocol.
>> Simple Reliability (SR) makes use of the FH to provide congestion
>> control and semi-reliability when using unreliable message-oriented
>> transport protocols."
>>
>> => In this and in followup sections, the operation on top of a reliable
>> congestion-controlled protocol (TCP, possibly SCTP in future) should be
>> better separated from datagram transport (UDP). The required functions
>> in the Overlay Link Layer seem to be quite different.
>>
>
> why?  e2e issues seem the same---even between consecutive TCP hops,
> the peers may experience congestion on the outgoing link and be forced
> to drop messages.

Ignore that comment, please.

Functionally, since TCP doesn't provide fast enough timeouts to be
useful as an overlay link protocol, an application-layer protocol is
required regardless of its underlying guarantees.
Implementation-wise, I think simplicity is the overriding concern.

Bruce
_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

Reply via email to