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 : LAST MILE TECHNOLOGIES - Let's Discuss Them Here -- Ignore unavailable to you. Want to Upgrade?

To: Frank A. Coluccio who wrote (4995)	8/18/1999 2:38:00 AM
From: Scott C. Lemon	Read Replies (1) \| Respond to of 12823

Hello Frank, > Scott, I sometimes (often) take your messages off line, as I did > this one, because your focus is somewhat off center to my center of > knowledge. A lot of people will say that my focus is just plain off center! ;-) > and I find that I must study them and occasionally look up some of > the references you cite. Your posts are always done very well, by > the way. I really appreciate that coming from you. I am always impressed with the broad base of knowledge that you demonstrate in your posts. I've learned a lot about the telco and WAN world from your writing! > In the process I sometimes don't get back to you in a timely > manner. Sorry 'bout that. Please ... no apologies. I keep wondering how you're about to read and write as much as you do! I often have trouble keeping up ... > Thanks for illuminating on this topic. I now have a better > understanding, although the dynamics behind what makes these remote > caches work are still evolving, it would appear to me, and by no > means have they found a final resting place in the greater scheme > of things. Yes ... I agree. I believe that they are going to become more and more a fundamental component of Internet infrastructure. As ISPs start to realize what they are capable of doing I believe that they will begin to create the parallel cached networks to today's packet networks. > I would imagine that here is a case where thorough cooperation > among service providers and total adherence to IETF RFCs must come > into play. Would you agree? Yes! Caches (and more specifically the proxy portion) rely on intimate knowledge of the higher-layer protocol that they are handling. One of the issues is that software vendors don't always implement protocols in exactly identical ways. The HTTP headers from Microsoft's Internet Explorer are slightly different then the HTTP headers from Netscape Navigator. The same goes for web servers. In some cases the RFCs have to be rev'd to make the necessary changes ... and in the mean time the proxy/cache has to implement some special case handling ... > More to the point, however, is the following question: How do ATHM > caching servers interact in the schemes that make your referenced > caches work? So this is where I'll admit that I don't find enough information. On the @Home site (http://www.home.net/about/network.html) there is some information about their network, which shows a very well thought-out solution. They definitely recognize the power of implementing a "parallel Internet" for performance and QOS reasons. But it also mentions that they are using "proprietary caching and replication technologies" and this is where I don't know just how proprietary they are. If they are based on the Squid research and technologies, then they should be somewhat compatible with the core HTTP and FTP protocols, and also compatible with ICP (Internet Cache Protocol) which is used for caches to communicate with each other. So the questions start to become: 1. Who's cache technology are they based on? 2. Are they able to integrate with "standard" caches? 3. What application protocols do they support today? 4. Are they able to handle multimedia streams and caching? What's interesting is that this is very parallel to router discussions. (i.e. what protocols do they route? IP, IPX, AppleTalk Do they talk standard route exchange protocols? RIP, OSPF, EGRP) I think they are in the "right" position in the Internet ... it's whether they have the right implementations. The "proprietary" word is always scary ... ;-) Scott C. Lemon Frank