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 : Cisco Systems, Inc. (CSCO) -- Ignore unavailable to you. Want to Upgrade?

To: The Phoenix who wrote (25859)	5/18/1999 6:36:00 PM
From: stockster	Respond to of 77400

Another interesting article in Communication news talking about PoS technology and ATM . INTERNET/IP TECHNOLOGIES Packet Over SONET Networking's packet-packed future Packet over SONET grabs the spotlight as a solo performer. by Mike Foley, Executive Editor With IP over SONET's emergence as a competent dance partner for--and competitor with--ATM in the backbone, information gets to experience a sort of "Dance of the Protocols" in the Netcracker Suite. It's interesting to watch the pair mix, while each also tries to show off individual moves and talents. ATM flaunts a certain amount of maturity onstage, an ability to perform well with multiple partners (voice, data, and video), and quality-of-service (QoS) delivery. But the ATM dance (although we are talking backbone speeds that outpace spins and leaps by light-years) is not without its latency-plagued missteps. The central problem is in transferring Internet protocol (IP) traffic (or any other frame-based protocol) to ATM cells and then back. Those two steps of conversion result in wasted bandwidth when compared to pure IP throughout the entire transfer. PICK YOUR DATA'S PARTNER WISELY "When you try to marry ATM and IP, you've got some troubles," says Yuval Boger, VP business development at RADCOM Ltd. This inefficiency is often referred to as the "cell tax," which is the overhead of various ATM layers that have to be converted. Inefficiency equals wasted bandwidth. And wasted bandwidth equals lost profits. Boger gives this example: "Let's say a typical IP packet for Internet browsing is around 576 bytes. If you transfer this over ATM on an OC-3 link, you are going to be able to use less than 80% (actually 79.6%) of your bandwidth for the actual IP transfer. The other 20% is overhead--ATM overhead or SONET overhead. Now, if you take the same packet and put it over Packet over SONET (PoS) technology, you find that you get 95.4% efficiency. So, over 95% of your bandwidth can be used for IP, and that's a big savings." RISING STAR As large enterprises, and ISPs especially, begin to deploy PoS in their backbones, the technology is not shy about showing off its best moves. "What I'm seeing is that many major ISPs either are using PoS today or they are moving their backbones to PoS," says Boger. "And this is Sprint, IBM, UUNET, and others. A lot of ISPs are saying PoS is the way to go." "Due to greater payload efficiencies, acceptance of Packet over SONET technology is booming," says Bruce Bowers, software development test engineer, optical internetworking business unit, Cisco Systems. It makes sense, for example, if you're an ISP delivering predominantly IP traffic (Internet browsing) that you would stick with IP from end to end. And ISPs especially may not need all of the advantages that ATM currently has for carrying voice. There is another difference between PoS and ATM, and it is one of complexity. When you're working with PoS, you're potentially using all of the same protocols that you used when you were just using IP routing. All the routing protocols in the IP suite carry over to PoS. In contrast, bringing ATM into the show requires a whole set of new controls for ATM that need to be dealt with and managed. "It is a very attractive proposition for an ISP to see end-to-end frames--not having to translate into any cells in the middle--and getting more efficient transfer," says Boger. "And since it is also a simpler technology, it will probably turn out to be cheaper. There are critics in the audience (there always are), but PoS can provide a markedly more efficient way of carrying certain types of traffic. One way to actually "see" the effectiveness of PoS and to maximize bandwidth use over large networks is through testing.

To: The Phoenix who wrote (25859)	5/18/1999 7:59:00 PM
From: zbyslaw owczarczyk	Read Replies (2) \| Respond to of 77400

Gary,in the real world there is need for different type of networks,like there is need for different types of cars. (Hmmmmm.... Go figure. A new network without an ATM core. In fact even worse...an existing ATM network migrating to packet over SONET) Many,specially corporste customers, where margins are high need QoS, reliability or provision of service like it is in the case of Korean network serving 500 corporate customers(see press release below) biz.yahoo.com The networks you are talking about is serving via cable average customers, who do not need high quality video-conferencing, do not care about private lines, banking transactions etc.What they want is fast and cheap service. Why IBM has chosen ATM form ATT solutions, which BTW is based on NN platform,for its new network serving many large US and international corporations. AT&T Solutions Managed Bandwidth Service is based on Newbridge Networks' MainStreetXpressTM ATM 36170 Multiservices Switch. brs.ibm.com 8?OpenDocument In 1998, AT&T Solutions received about $7.5 billion in contracts for new long-term business -- including a $5 billion contract from IBM, a $1.4 billion contract from Bank One Corporation and a $750 million contract from Citibank. More information about AT&T Solutions is available at www.attsolutions.com. Regards Zbyslaw

To: The Phoenix who wrote (25859)	5/18/1999 9:30:00 PM
From: bill c.	Respond to of 77400

>> Could this be what Cisco was thinking when it killed the TGX??? << No.... They killed the TGX project from lack of resources, competition, lack of market share and they couldn't develop a cost effective ATM core. Lets face facts... they failed to deliver a product and they are late on other ATM products, now they are promising a new "Juniper" ATM/IP core in late 2000. I'm really enjoying the Gary-spin-master at work Keep on spinning, you may convince yourself this wasn't a failure.