Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : The *NEW* Frank Coluccio Technology Forum -- Ignore unavailable to you. Want to Upgrade?

To: axial who wrote (3264)	7/15/2001 4:05:50 PM
From: axial	Read Replies (1) \| Respond to of 46821

Frank (just missed the late edit!) - another contribution to latency, interesting in this 'convergence' discussion - is the delay introduced by fiber, as opposed to copper, transmission. Trivial over 100 Km., but it becomes significant at 10,000 Km. Jim

To: axial who wrote (3264)	7/15/2001 5:49:14 PM
From: Frank A. Coluccio	Read Replies (1) \| Respond to of 46821

Hi Jim, I've been meaning to revive the VON thread lately, because I think we will soon be nearing the stage where wide scale commercial viability may be approaching prove-in (or bust). In the 1997 time frame I was very active there, in no small way due to my growing interest as a principal in a development stage Internet Telephony SP (ITSP) that was striving to do something along the lines of Tom Evslin's ITXC, prior to knowing about his venture. Fortunately - and I do mean fortunately - our venture disintegrated after about a year under its own pressures, and the realization that there was virtually no barrier to entry. Whew! That was close ;) In any event, many of my posts on the VON thread during that era were evidence of the levels of frustration that many other startups were going through, since the voip framework was not, as many had envisioned, a straightforward erector-set affair. Every other week a new protocol suite was being introduced into the mix, obsolescing the previous gateway designs, which were based on cruder form of discreet component makeup, and not the ASICs and DSP chips that are now commonplace. One such frustration was getting some of the top tier vendors to agree that SS7 would play into the overall voip gaeway framework, in some manner. After going to LU and CSCO earlier with no satisfaction, I found myself going to Excel Switch and Summa Four. I had each of these outfits come to our offices, and they didn't know what I was talking about at first. Turns out, CSCO acquired SUMA and LU acquired Excel Switch, both for the same reasons that I was pursuing them in the first place: for an integrated IP-SS7 solution. And guess what? That was only a short-lived solution for those vendors, at best, since today SIP and softswitch have taken over this space, in spades. Enough lamenting... --- Getting to your questions: "What I had hoped was that we could abstract part of the answer: i.e., given an otherwise 'perfect' system, with no capacity constraints, what portion of the latencies could potentially be eliminated by IPv6? My understanding of the intent is that ultimately everyone would be reachable by their network address, as opposed to reliance on lookups at the CO, for instance." Given that v6 actually carries more overhead than v4, I don't know, to tell you the truth. If you question centers on the avoidance of lookups in directories (NAT and DHCP translations due to the use of unique addresses, enabling end-to-endness without the need for leasing addresses/proxies), then I suppose there would be some improvement, but I could only guess as to how much, or if it's appreciable. The primary cause for latency these days, however, is hop count and ordinary router lookup times at each node along the way. "in such a 'perfect' system, could IPv6 enable "shortest route" switching for VON?" I don't think v6 addresses that capability in ways that are materially different than v4. Instead, shortest route might be achieved through traffic engineering and through the use of MPLS, with supports in place for differentiated services. "Maybe the issue is so complex that no sensible answer is possible; if so, my apologies." No need for apologies. It's been a while since I've examined the voip space, so I'm as interested as you in hearing opinions on these issues from others who lurk here, who I know "are" experts in IP routing and VoIP. Maybe we can get an assist here from ftth, Darren, John Cavanaugh, Curtis Bemis... there's a whole bunch of talent out there who can help answer your questions. FAC