End-to-end arguments in system design
ACM Transactions on Computer Systems (TOCS)
On characterizing BGP routing table growth
Computer Networks: The International Journal of Computer and Telecommunications Networking - Special issue on The global Internet
No Administration Protocol (NAP) for IPv6 Router Auto-Configuration
AINA '05 Proceedings of the 19th International Conference on Advanced Information Networking and Applications - Volume 2
HAIR: hierarchical architecture for internet routing
Proceedings of the 2009 workshop on Re-architecting the internet
A secure mechanism for address block allocation and distribution
NETWORKING'08 Proceedings of the 7th international IFIP-TC6 networking conference on AdHoc and sensor networks, wireless networks, next generation internet
| Hi-index | 0.00 |
In order to suppress the increase in the global routing table size, address hierarchicalization and multihoming that does not depend on routing mechanisms are required. This type of multihoming can be achieved by a method that allows an end site to be allocated multiple address spaces from upstream ISPs and utilize all of them. We call this type of multihoming end-to-end (E2E) multihoming. In E2E multihoming, when a site adds or changes an upstream ISP, the addresses are also added or changed. If a network manager must handle these changes manually, these changes become too cumbersome a task for him or her. Thus, in order to deploy E2E multihoming, a hierarchical automatic renumbering protocol is required. We propose a Hierarchical Automatic locator Number Allocation (HANA) protocol, which considers hierarchical end-to-end multihoming, and reduces the manual router settings. In the HANA protocol, the lower part of an addresses (midfixes) are automatically allocated in each domain, and the upper part of an addresses (prefixes) are distributed from the upper-level ISPs in multiple layers, in the manner as the previously-allocated midfixes are combined. In addition, we discuss the implementation and evaluation of the HANA protocol, and show that it is manageable in real networks.