It was my (possibly erroneous) impression that HTTP/1.1 was to get all
objects using a single TCP connection rather than using one per object
in serial or parallel. Is that not the case, or is the problem involving
an excessive number of legacy HTTP/1.0 clients and/or servers?
Matt
> -----Original Message-----
> From: Sean M. Doreen [mailto:s...@clock.org]
> The problem is that the large source in question is sending
> multiple parallel flows along the same part of the tree, and
> two things will tend to happen:
>