Terena Title Logo  
decoration line

navigation button: home navigation button: programme navigation button: meetings navigation button: social navigation button: registration navigation button: venue navigation button: credits navigation button: contacts navigation button: sponsorship navigation button: search
decoration line

TERENA Logo small


GÉANT 2 : A New Approach to Networks

For the past ten years, the approach to pan-European research networking has been simple; it has been to build bigger and faster IP networks. This simple strategy has been significantly helped by the fact that the cost of connectivity, which has historically been the dominating cost of European research networking, has dropped dramatically in this period. Thus, the basic building blocks of pan-European research networking have evolved from 2 Mbps PDH connections to 10 Gbps DWDM connections. Looking ahead, it is apparent today that the successor network to GÉANT will not be a simple extrapolation of this approach. There are at least three factors that demand a new approach in terms of the way the new network will be built and operated. Looking at these in turn:-

Cost of Connectivity:
Although connectivity costs have dropped dramatically in many parts of Europe, a trend that has continued since 2000 despite difficulties in the telecommunications market place, it is now the case that there are huge variations in the underlying cost of connectivity across Europe. The way costs are shared in GÉANT has done something to mitigate the effects of this digital divide but a more direct solution is necessary. The potential to operate and implement "owned" connectivity in the context of GÉANT 2 will lead to new approaches to the provision of service.

Evolution of User Demand:
The availability of very large network capacities, on a pan-European basis, is starting to give rise to new usage of a network which involve very large point to point flows of the order of hundreds of Mbps to Gbps. These are in marked contrast to the historic usage patterns where large flows are typically made up of multiple smaller flows aggregated together. Experiments with radio astronomers have shown that some applications have requirements for very large bandwidth, but also that these flows do not simply happen. They need to be supported and engineered.

Hardware Costs:
Whereas in the past connectivity costs used to dominate, now, as speeds of operation have increased, the cost of hardware to interface connectivity is becoming increasingly significant compared with the cost of connectivity itself. Whereas connectivity costs generally show some reasonable economies of scale with increasing speed this is not the case today for hardware costs. In the case of hardware, it appears to be the case that costs increase annually with speed of operation.

All of these factors contribute to the need for new and different approaches to be adopted in the successor network to GÉANT. This presentation will review how the next generation of network will develop and speculate about its likely form from a service and from a topology perspective.



This presentation is part of session "Rollout I" which starts at Tuesday, June 8 @ 11:00


Home | Programme | Meetings | Social | Registration | Venue | Credits | Contacts | Sponsorship | Search back to top
Last modified on the 15th 2004f June 2004 - 12:35