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 : Voice-on-the-net (VON), VoIP, Internet (IP) Telephony -- Ignore unavailable to you. Want to Upgrade?

To: LAWRENCE C. who wrote (693)	6/4/1998 1:40:00 AM
From: Scott C. Lemon	Read Replies (1) \| Respond to of 3178

Hello Lawrence, > Scott, a key point you make is that you measure the impact of > changes you make by adding the cache. Yes, there are a variety of aspects being measured today. Some of them are related to bandwidth in and out of the cache, so that an accurate determination can be made as to whether the cache eliminate duplicate traffic ... this would provide more available bandwidth for applications (such as VoIP) which require end-node to end-node communications with a high QOS. The other measurement is to determine the end-user response time when viewing content or downloading files. We have seen a big increase in the number of pages delivered to browsers per minute when using a cache. This does bring up an issue that Frank mentioned ... in one set of tests we found that installing the cache did not decrease the bandwidth utilization of the Internet connection ... but this was because the users were able to "cruise" 10 to 100 times the number of pages per minute! Again ... give the user more "virtual" bandwidth and they will use it up! ;-) > The carriers will measure the amount of traffic over their nets. Exactly ... the current model of "unlimited" Internet access for applications I believe will stand ... but only on dial-up access. The user will always be throttled to 56kbs max. But with the new services such as ISDN or xDSL there is already a model that appears to be forming which charges for the purchased bandwidth. > Individual subscribers will buy either a certain amount of bandwidth > or bandwidth on demand according to some schedule. Customers will > have levels of service available to them. This might be in the form > of a dedicated level of service or bandwidth on demand. There are > many private nets that handle data already. VoIP will be another > feature. I agree ... I think. Users will have the option to purchase bandwidth, and I believe to request prioritization on demand ... these capabilities are being built now and are based around directory services and proxies which can throttle based on protocol types. I think that some innovative ISPs will implement systems which will throttle some protocols if they are used through traditional IP routed networks. The user will be given better performance and less throttling if they are redirected through caches. HTTP and FTP for example would do better through caches, VoIP would go through the traditional IP routed network. I don't think that VoIP is going to end up being a special protocol or service ... I use it today (even with video) over 56kbs dial-up and higher bandwidth links ... I don't need to pay extra today and I don't think I will in the future ... > MULTIMEDIA DATA: Multimedia already flows over many data networks. Yep ... I listen to an 80kbs audio stream most days while at the office ... CD quality audio delivered via RealNetworks from a variety of web sites. This is over a T1 with an 800kbs CIR. > Although sometimes it is not handled well. When my segment of the > LAN saturated the repeater I had a big problem. (I've managed LAN > segments that were part of large networks.) All the users in > my building could not use that ethernet cable to connect to other > segments of the LAN because the traffic was more than the repeater > could handle. We had to add a bridge before the repeater to reduce > the amount of traffic coming across to the repeater in our building. > Managing traffic growth is a major part of a LAN/WAN manager's > responsibiles. One way we handled high demand traffic flows was to > put high demand local traffic flows onto their own segment of the > LAN. And then routers were invented and started to permeate the infrastructures of LANs ... and routers further segmented the LANs to control unnecessary traffic from spreading to segments that were not part of the communications. And then "layer 3 switching" arrived which is simply a way to accelerate the routing process ... but it does not address some of the issues surrounding traffic which does not need to leave the local LAN. By making a "router" intelligent about the protocols and the "cargo" being carried we can eliminate much of the redundant traffic. This leaves more bandwidth for: > VoIP > Handling VoIP is what's relatively new. Well ... only in a more commercial implementation. > Home User Market: > Essentially the home user has many level of service related choices: I now use a cache at home on a cheap clone PC due to the considerable performance gains. And I have tied my cache into a hierarchy with the cache at my ISP to allow the two caches to work together to remove latency and maximize the use of my dial-up bandwidth. > In the case of a cable user, the cable company would have to decide > how to allocate it's bandwidth. But in any case providers will have > to monitor traffic flow and watch for bottlenecks as the ISPs have > to do now. Many of the internet problems are bandwidth at the > internet site problems. Popular communication sites with a large > number of users like AOL or SI require a lot of bandwidth. Yep ... and AOL is a big user of caching to relieve infrastructure bandwidth problems. They fully recognize that much of what the AOL users access is highly redundant. I agree that the ISPs like cable companies are going to have to be careful about giving out bandwidth without seriously thinking about how to control the flows ... > Digital PCS > Perhaps VoIP companys will offer companies with digital PCS low > international calling rates to/from it's customer's cell phones. It > will collect the monthly fees plus additional revenue for > the low international calling rates. Then you don't have a pipeline > problem to the customers for voice. It is common to see people using > cell phones in malls or car phones. As Digital PCS usage costs > drops, Digital PCS becomes more attractive to more users. In this > niche you already have fraud management at that level. Just switched to AT&T OneRate Digital PCS ... this is an amazing service. To me the big gains will be from the area that Frank discussed a while ago ... the bridging of legacy phone systems to cellular, and to pure IP software phones that are running on PCs. ;-) > In any case we will have a hybrid network of VoIP communication and > traditional data communication links. I guess that I look at VoIP as traditional data. It's just a stream of data that needs a little bit of special treatment. This can be handled in a variety of ways such as building the infrastructure to prioritize the traffic, and also developing technologies which offload unnecessary traffic from the links. P.S. I don't think VoIP is anything big ... IMHO it will be still-born like ISDN ... yes, we'll see it in someplaces and maybe in some fairly significant numbers, but I'm using video over the 'net today ... I think that the long awaited "VideoPhone" is just around the corner! > Lucky Lawrence Scott C. Lemon