In article <MPG.1d80c37d47f99e899896b2@newsgroups.comcast.net>, Claude J
<cj-usenet-01_spam@comcast.net> writes

Berkeley turned the Classic servers off for a while, to allow the Boinc/Seti 
server to handle the surge.

Unfortunately this does not seem to have released the capacity needed. 

In the past, after an outrage on either Classic or BOINC, the first few
hours on the available bandwidth has hit a sustained peak of 85-90 Mbps.
This time around, with the Classic servers off, the rate seems to have
peaked at about 45 Mbps. This has now reduced to circa 30 Mbps.

With the sheer SETI-BOINC demand & numbers trying to upload results /
download WUs after a 7 day outrage, the failed handshaking will be
stupendous (acknowledged by BOINC in yesterday's Tech News).

The full bandwidth would help, and clear the demand more quickly than
what seems to be a capped bandwidth as currently prevailing.

Looking at the bandwidth graphs at -

        http://fragment1.berkeley.edu/~cricket/inr-668-interfaces.html

- the second graph down (left hand side) shows the usage of a 100 Mbps
bandwidth capacity to be 45 Mbps at midnight Tuesday/Wednesday (UTC).
This despite the Classic servers being shutdown to remove this demand.

Currently the demand seems to be dropping towards 25 Mbps, and the slope
is quite clear over the last 36 hours.
-- 
Hugh Janus      Constipation is the thief of time, but diaorrehia waits
                for no man!!