Qofis'2000 Papers

M3I Newsletter 

Subscribe to M3I 
Newsletter (free)

Private Server

The goal of this project is to design, implement and trial a next generation system which will enable Internet resource management through market forces, specifically by enabling differential charging for multiple levels of service.  Offering this capability will increase the value of Internet services to the customers through greater choice over price and quality and reduced congestion.  For the network provider, flexibility will be improved, management complexity reduced and hence revenues will increase. 

This project takes a radically different approach to differentiated services.  Its premise is that a simple packet network may be able to support an arbitrarily differentiated set of services by conveying information on congestion from the network to intelligent end-nodes, which themselves determine what should be their demands on the packet network.  There would then be no need for large buffers or priority queues within the network, or connection acceptance control at the border of the network. 

Measurable Results

A trial system will be designed, built and experimented with.  It will enable ISPs to explore sophisticated charging options and business models with their customers. 

Measurable improvements that an end user of our system will experience are:

• The ability to instantaneously increase QoS by accepting different charging rates;
• More effective competition in a differentiated services market;
• Real-time feed-back and validation of charges.

• The ability to change tariffs and easily communicate them to the end users within seconds;
• The reduction of bad congestion effects by communicating price changes in real-time to customers;
• The ability to charge differentially for applications requiring differing QoS levels, or multicast.

• The demand for Internet services, including various Quality of Service (QoS) levels can be managed effectively through a pricing mechanism;
• Customers can flexibly access both high quality and low quality services, depending on their particular application needs, instead of being limited to a single best-effort service as in the current Internet;
• End-users in corporate organisations can exercise similar choice, but constrained by the policy of the party that is paying;
• Internet providers can recover the costs of new services, such as voice and video, that are currently provided by different infrastructures, and hence increase both social efficiency by exploiting economies of scale and multiplexing, which in turn will also provide for increased network revenue;
• Simple and scalable extensions to current technology can provide the correct incentives for the economically efficient and uncongested operation of the Internet.

Outline

We believe that the present position whereby Internet Service Providers charge a flat rate for access to undifferentiated service levels and meet demand through over-provision of capacity, does not represent a sustainable business model.  As basic access and simple services become commoditised ISPs will differentiate across a wide range of sophisticated services and business models.  This project will design and prototype a system that can charge and account for multiple levels of Internet service, and analyse the implications of so doing.&nsbp; We believe that this is critically important for the industry. 

We intend to demonstrate the feasibility of managing the allocation of Internet resources using relatively direct market forces.  Feasibility will be established in technical and commercial terms.  This includes requirements analysis, engineering design and prototyping, cost and business modelling, and Internet market modelling.  The result will be a fully operating prototype that will enable market trials. 

By Dec 2001 (the end of the two-year project) we will demonstrate relatively sophisticated applications reacting to economic signals from the Internet and report our understanding of typical customer reaction and the effect on total system performance. 

Using our results we will:

• Demonstrate that any user of the Internet can make highly serendipitous bursts of usage without compromising the right of any others to do the same, as long as all are prepared to pay a premium if congestion results;
• Examine whether a multi-service Internet with decentralised charging is more cost-effective than simple over-provision of network capacity billed at a flat rate;
• Demonstrate that users' sensitivity to increases in price at times of network congestion can be the basis of effective load management.

M3I is RTD Project No IST-1999-11429 under the European Union's Fifth Framework Programme.  It is part of Key Action 4 "Essential Technologies and Infrastructures", Action Line 4.2.4 "Technologies for network management and service level interworking".  The total of eligible costs for the project is €3.7M, with a maximum community contribution of €2.1M.  The participation of ETH Zьrich is supported by funds from the Swiss Confederation.