The users of the net want flashy web pages and better performance. They tend to use sites that visually excite them and are respond quickly, this is human nature. All commercial sites want a competitive advantage, such as speed. Continuously customers are told "Buy more bandwidth if you want better performance." Yet link utilization rarely goes above 50%(in my case). ISPs and NSPs say that the problem is overloaded web servers, yet processor utilization has never been lower(<10%) as more web servers are upgraded to latest Pentium. Complex HTML from our local boxes pops up on our screens here, but try it through our customers ISP. The lease line users want answers. The dialup customers want answers. Both want to use networks that perform faster. They want to know where the problems are so that they can avoid them. If Sprint has an over utilized T1 to CIX, I want to know. I will let my local engineer and sales rep know where the T is and if Sprint will not fix it maybe it is time for me to try a non big name backbone provider. I think that the MAEs and NAPs should publish utilization of all ports and that NSPs should publish link utilization to interconnects as well as long term plans for their networks infrastructure. And router CPU stats. Boardwatch is giving end users what they want some guidance and direction. This study is flawed in many ways. I think that a study of performance to the top 50 websites and the top 50 ISPs from both Dialup and leaseline with a web browser and A FTP client would be very interesting. Does management want to hide performance figures? They seem to want it to be a PR and Advertising war. NetNow had great graphs of average delays and performance. Why is this information now unavailable. I guess since I don't have a nation backbone, I can't understand a simple graph of performance. Maybe NETNOW should run in Latin. BTW I don't want to spend 50,000$ to find out what is wrong with your networks. I speak for end users wanting better performance. Please tell us where you see problems on your network and others. Remember, The better you make the net work the more data we will push through it and the more you can sell us. Peter Cole of Telescan, Inc. (281)588-9155 Better computing through lack of sleep.
---------- From: Mike Leber[SMTP:mleber@he.net] Reply To: Mike Leber Sent: Thursday, July 10, 1997 3:55 PM To: Gary Zimmerman Cc: Sean Donelan; nanog@merit.edu Subject: Re: Internet Backbone Index
You talk about engineering good networks, why then use something
On Thu, 10 Jul 1997, Gary Zimmerman wrote: that is
unmanaged and uses poor technology and poor engineer designs in a MAE or NAP? Where is most of the packet lost on the internet? MAEs and NAPs.
This is gratutitous and misleading to the journalists and other non-network engineers here.
The packet loss at NAPs and MAEs is caused by providers not upgrading the size or number of connections they have to/from the various exchange points they are experiencing problems at.
An example of this disingenous usage of exchanges is Sprint's completely saturated T1 connection to CIX. There are many other examples. The solution is to not use those routes.
If a provider is leaking bad routes and flaps continously, would you peer with them? Likewise, if a provider has a notoriously saturated connection to an exchange does it them become a (misleading) casebook example of why exchange points are bad?
No and No.
Mike.
+------------------- H U R R I C A N E - E L E C T R I C -------------------+ | Mike Leber Direct Internet Connections Voice 408 282 1540 | | Hurricane Electric Web Hosting & Co-location Fax 408 971 3340 | | mleber@he.net http://www.he.net | +--------------------------------------------------------------------- ------+