Architectural analysis and instruction-set optimization for design of network protocol processors
Proceedings of the 1st IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis
Design of an efficient memory subsystem for network processor
Proceedings of the 2005 Asia and South Pacific Design Automation Conference
Autonomous agents for self-managed MPLS DiffServ-TE domain
AN'06 Proceedings of the First IFIP TC6 international conference on Autonomic Networking
| Hi-index | 0.00 |
Over the next 10 years (2000-2010), a lot will happen in multiple technology areas (high-speed access to the home, high-speed applications, user-friendly applications, and security). We will see a huge increase in the demand for bandwidth, which will drive major changes to the high speed core of the network. Bandwidth will continue to grow rapidly. The rate of growth will average a factor of 4 per year over five years (implying total growth will be roughly a factor of 1000 relative to 1999 bandwidth). After that, growth might abate, but it will still be at least a factor of two every year through 2010. Optical switches with connection-oriented control planes will be used in the core because they provide very high bandwidth and density at a reasonable cost. Very large IP routers will, however, be necessary for statistical multiplexing and ubiquitous any-to-any connectivity. The core will therefore consist of multiterabit or petabit routers interconnected by very high density optical switches. As networks and bandwidths grow, it will become impossible for humans to do traffic engineering and to provision SLAs (service level agreements)