Here I'll try to collect my current experience in ATM networking. As it
is, I have only tried Classical IP over ATM.
It all started with me getting involved in the MERCI project in early 1997
where I was supposed to do a Multimedia Server survey, but when the new,
hot ATM switch that we used in our corner of the
MERCI high-speed MBone decided to give up, someone had to fix it.
And suddenly I was the local expert on ATM...
This high-speed Mbone consisted of a number of PVC's over the JAMES testbed and since the whole configuration was gone with the switch, I had to figure it out using an old documentation made by Magnus Danielsson the last time he switched switches. Talk about crash-courses in ATM, all you need is 20-pages of old configurations a few configuration manuals and six months of study time. Ofcourse I didn't have so much time to fool around with, so we got help from Bengt Ahlgren at SICS and he gave us a crash-course in ATM (less than an hour) and then we switched back to our old switch and Bengt configured it and everything worked fine for almost 42 hours until the hard-disk gave up in the new (old) switch...
In pure desperation, Bengt then went up to SICS and fetched their switch, sidetrack, and hooked it up and configured it and it worked fine for a couple of weeks until our link to JAMES suddenly became one-way only. By then I gave up on ATM completely. However, the final setup looked like this .
Now the MERCI project has ended and the network activity with it. Instead I got involved in another project (again in the role as ATM expert), the Graduate School in Teleinformatics. They had a problem. The node in Högskolan Karlskrona- Ronneby had leased an ATM link and wanted to plug it in somewhere on our side. In the flurry of activity, I suggested the following solutions:
The 3rd alternative was the one they decided to use since sidetrack ( the ATM switch that we've borrowed from SICS) still hadn't been replaced (someone decided to scrap our switch some 30 000 feet over the altlantic on the way to service in US). They say we'll get a replacement, but so far no see...
Maintained by Tobias Öbrink
