TCP Daytona in userland

[Mark Allman]

> Lastly, if anybody already worked on this: do you have any additional
> suggestion? In particular regarding the testing methodology: since a
> few years have passed, I'm not quite sure whether different OSs have
> implemented any countermeasure. I'm mainly testing against a FreeBSD
> box I control, and I don't think we have any defence against this yet.

It would seem that some hosts are using byte counting to increase cwnd
these days (that is, increasing cwnd based on the number of bytes ACKed
and not the number of ACK packets that arrive).  There are some
measurements given in:

  Alberto Medina, Mark Allman, Sally Floyd.  Measuring the Evolution of
  Transport Protocols in the Internet. ACM Computer Communication
  Review, 35(2), April 2005. 
  http://www.icir.org/mallman/papers/tcp-evo-ccr2005.ps

Also, I think there is wide community consensus that cwnd should be
increased by min (number_of_bytes_acked, MSS) on each incoming ACK.
RFC2581 is currently being revised and this will be the recommended way
path in the revision (in 2581 it notes that an implementation may count
bytes instead of packets).

allman



-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-net/attachments/20060411/3d5d5b89/attachment.pgp