Peter, congratulations again. - Could anybody help me a bit with SONET/SDH technology: what is the present state of - say STM4c vs STM4 etc.- standardization. At Stuttgart we run some 20km dark fibre driven by FORE`s ATM/STM4c boards (NEC SX4 32 - Cray T3E 512) - would we buy e.g. NORTEL SDH equipment if this would be - say 100km - and what about public European SDH? How is vBNS going to upgrade to 622Mbit/s ? What does the the below "Cisco's Native IP over SDH/VC4" exactly say? Paul ---- p.s.: ... if not the telephone system itself- would Christian prepare an analysis saying "a 600Mio participants Internet of `appropriate` capacity with RSVP and IPng would allow for concurrent AUDIO (as Jon said) of phone quality for everybody ..." :-) ------------------------------------------- On Dec 2, 2:28pm, Peter Lothberg wrote:
Subject: Re: Very fast IP, no ATM...
p.s.: have you any measures of IP-level throughput over this link?
Almost 2 moths of operation, 4h downtime and we have been running upto 120Mbit/s user traffic.
(Traffic generated by traffic generators does not count, but it's possible to fill it up to theoretical max.)
It's just a releif to see this first link that have enough buffer memory to deal with the network burstiness and allow TCP to do it's job..
--Peter
On Monday 23 September 1996 00:15Z..
Was the worlds first transatlantic 155Mbit native IP service brought into operation by Sprint/USA and Tele2/Sweden.
The circiut wich is part of Sprintlink and ICMnet runs between the NY-Nap in Pennsauken, NJ, USA and Tele2 in Sweden and uses cisco packet-over-sonet/sdh technology. (Native IP over SDH/VC4) -- End of excerpt from Peter Lothberg
p.s.: ... if not the telephone system itself- would Christian prepare an analysis saying "a 600Mio participants Internet of `appropriate` capacity with RSVP and IPng would allow for concurrent AUDIO (as Jon said) of phone quality for everybody ..." :-)
weel, for toll quality speech, we can get down to around 16kbps per user, i.e. 4 per 64kbps, i.e. nearly 100 per 2Mbps, or 30,000 per 622Mbps... now assume that out of the worlds 600M phones, 10% are active and assume that most calls (say 90%) are local then we need to tolerate a total of 6M calls....(this is WAY higher than they actually do...) assume that we build a path disjoint backbone, so continental locality is also kept reasoanbly well (and timezones help anyhow) - we need to support 600,000 calls on an Internet trunk... so we are out by 20 times right now..... ok, so we put in a 16 fold replicated backbone global grid...then we're there but there are 800k web browsers in the UK after our interntational links anyhow, so we need to upgrade anyhow... so they'd all suddenly get instantaeous reponse... oh, the cost? tricky one:-)
------------------------------------------- On Dec 2, 2:28pm, Peter Lothberg wrote:
Subject: Re: Very fast IP, no ATM...
p.s.: have you any measures of IP-level throughput over this link?
Almost 2 moths of operation, 4h downtime and we have been running upto 120Mbit/s user traffic.
(Traffic generated by traffic generators does not count, but it's possible to fill it up to theoretical max.)
It's just a releif to see this first link that have enough buffer memory to deal with the network burstiness and allow TCP to do it's job..
--Peter
On Monday 23 September 1996 00:15Z..
Was the worlds first transatlantic 155Mbit native IP service brought into operation by Sprint/USA and Tele2/Sweden.
The circiut wich is part of Sprintlink and ICMnet runs between the NY-Nap in Pennsauken, NJ, USA and Tele2 in Sweden and uses cisco packet-over-sonet/sdh technology. (Native IP over SDH/VC4) -- End of excerpt from Peter Lothberg
jon
Jon Crowcroft writes:
weel, for toll quality speech, we can get down to around 16kbps per user, i.e. 4 per 64kbps, i.e. nearly 100 per 2Mbps, or 30,000 per 622Mbps...
now assume that out of the worlds 600M phones, 10% are active
This might be realistic but is actually on the higher side.
and assume that most calls (say 90%) are local then we need to tolerate a total of 6M calls....(this is WAY higher than they actually do...)
Exactly. The average active international call count for 3 million finnish telephones is less than a thousand. That would make the international call percentage 0.033% of active call minutes which sounds realistic.
assume that we build a path disjoint backbone, so continental locality is also kept reasoanbly well (and timezones help anyhow) - we need to support 600,000 calls on an Internet trunk...
And with the above more realistic figure, this number comes down to 20000. Which in order sounds still high knowing the infrastructure in existence to carry the calls.
so we are out by 20 times right now.....
Take a short look how many bits per second is actually in use between the continents and then figure out where the phone calls must fit :-)
ok, so we put in a 16 fold replicated backbone global grid...then we're there
but there are 800k web browsers in the UK after our interntational links anyhow, so we need to upgrade anyhow...
so they'd all suddenly get instantaeous reponse...
oh, the cost?
tricky one:-)
When we get more privately funded intercontinental cables like FLAG, the availability of bandwidth for a reasonable price will be here. Until then, we've just wait for the internet-aware cable investors to pop up. Note that it takes half a decade from a plan to a cable. Pete
participants (3)
-
Jon Crowcroft -
Paul Christ -
Petri Helenius