A couple of comments on this thread...
1) I ran into this problem when I was running the web server at Purdue.
At one point we were accounting for 20% of all the traffic on the
external T1s. At the time, there was a BIG debate over what was
considered justified use of Internet bandwidth. At that time, web
servers were not considered justified and there was a solid possibility
I would have to shut down my web server. As a result of this, Purdue
was contemplating charging for bandwidth usage. The one likely scenario
was to charge departments only if they exceeded a specific bandwidth
limit in terms of megabytes per week. This limit would be set based on
average traffic levels. So for email, research related FTP, etc, you
wouldn't get charged extra. You still had port connect fees. If you
had a web server with a lot of traffic (or IDD), it would essentially
act as a bandwidth fine. Luckily for me, this type of monitoring was
It sounds like issues over paying for bandwidth are being reintroduced.
This doesn't surprise me as universities are spending more and more
money on Internet bandwidth. Its not like the cost of a Mbit/sec in
bandwidth has gone down in recent years.
2) As for NOAAPORT systems, the costs of these systems should come down
with time. We continue to investigate newer, faster, cheaper hardware
for accessing the NOAAPORT feed. The computer hardware required to
handle the feed is getting cheaper. Even the costs of OS software, now
that Solaris is essentially free, are coming down. But will this make
commercial NOAAPORT systems cheaper?? I don't know. Since I'm at
Unisys and we have a commercial NOAAPORT system, I can't really comment
or justify the costs of our NOAAPORT system. But the costs are higher
than what Mike Dross detailed. The biggest issue not listed in his
price list is support. We've found this to be a large cost especially
in the commercial world where knowledge of weather data and
communication hardware is relatively limited. We have to pass those
costs on to the consumer. This may be different in a university
environment where these support costs wouldn't be as high, thanks to
Unidata and others like those on this list.
As for NOAAPORT vendors, I know Planetary Data and Unisys offer LDM
solutions. The cost of each setup is roughly the same but not having
worked with the Planetary Data system, I can't detail advantages or
disadvantages of each system relative to the other. If I remember
correctly, their system is a Linux based system whereas ours is Solaris
x86 based. I'll let others debate the merits of each system.
If anyone has questions about NOAAPORT systems, I'll be glad to answer
what I can.
Daniel Vietor Mail: devo@xxxxxxxxxxxxx
Unisys Corp Title: Engineer/Meteorologist
221 Gale Lane Phone: 610-925-5206
Kennett Square PA 19348 Fax: 610-925-5215