CISCO meeting: Future of the Internet

The future of the Internet: what's keeping the CISCO engineers awake at night.

I attended this CISCO presentation to the University of Oxford today.

The first presentation was from Fred Baker (CISCO fellow and ex-IETF chair).
Talk focused on growth of INternet and current limit of number of computer addresses. The goal is to, "Continue the growth of the Internet and its businesses"; for CISCO the goal is, "Continue the growth of the Internet with maximised application options and minimalised additional cost."

RFC521 - John Curran's Internet Transition Plan for ISPs: explains transition from IPv4 to IPv6. Fred explained how the two systems would have to co-exist.

CIDR put off the need to go to IPv6 to deal with the shortage of addresses.

A particular issue is the amount of information that is needed in the routing if you put an address on everything you want to track. IPv6 routing scales better than IPv4.

There are serious IPv6 trials in enterprise services. Total traffic is still less than 1% of total traffic.

The second presentation was from Klass Wierenga (consulting engineer in the office of the CTO) entitled the Mobile Internet

Billions of new users onto the Internet, most mobile access.Tens or hundreds of billion things connecting.

Radio access: RF is a major bottleneck - does not scale.

• Need to optimise airlink efficiency.
  
• Need to make networks a lot less expensive to build and operate on a cost per bit basis.
• Need to optimise routing; possibly not use tunnels, expose mobility to hgher layers and make it clear that host is moving...

Who decides who gets access to a network? Likely to require some kind of roaming agreement and technology for remote authorisation.

There are three identity spaces: SAML, OpenID and CardSpace. Trust is the foundation of any security model.

Network access: eduroam: authentication by home institution, authorisation by visited institution.

Application access: Shibboleth: UK Access Management Federation for Education and Research. Developed by Internet 2, uses SAML.

Conclusion: The issues of the Internet relate to scalability.

-----------------------------------------------------------------------------------------------------------------------------

As go through transition, Internet industry need a lot of help from academia. Internet is an open confederation and so everyone needs to contribute to evolution.

Is it true that the Internet has done as much good as it has done bad (a statement made at the meeting)?