A Method for IPv6 Sitelevel Address Allocation
LIU Lan
(Hangzhou Dianzi University, Hangzhou, 310018, China)
Abstract:The disadvantages of IPv4 Address allocation make IPv4 network ineff icient and address space wasteful. IPv6 address allocation must be managed in a prudent manner with regards to the longterm interests of the internet fo r imple mentation IPv6. This paper proposed a method for IPv6 sitelevel global unicast address allocation which is followed the latest RFC documents of IPv6 addressing architecture and IPv6 allocation policy. The proposed method not only has flex i ble scalability but also supports different policy for different service. It is a foundation of implementing IPv6
Keywords:IPv6; addressing architecture;address allocat ion policy;scalability
在开展IPv6实验和应用时,首先要做的是IPv6地址分配。虽然IPv6有巨大的地址空间,但吸取IPv4地址分配的经验教训,以谨慎的方式合理地分配IP地址是非常重要的。提高路由性能又不使路由表过于膨胀是地址分配要达到的一个目标。世界各国同盟在这方面做了大量的工作,从1995年12月的RFC1887(an architecture for IPv6 unicast addres s allocation)开始,到2004年6月发布的RFC3769[1],其间推出了几十篇有关的RFC文档和草案。新的方法和原则不断取代旧的方案,使其更加适应当前和可预期未来的需要。随着地址分配经验的成熟和技术的提高,这个过程还将不断进行下去,直至找到一种能满足所有需求的完美解决方案。
IPv6单播地址可用任意比特长度的前缀来聚合,这一点与IPv4的无类域间路由(CIDR)相 似。图1为RFC3513[2]定义的IPv6全球单播地址的一般格式:这里全球路由前缀一般具有分层结构,由RIR(Regional Internet Registry)和ISP(Int ernet Service Provider)将其分配到站点Site(一族子网或链路);子网ID是站点内一个子网的标识符,他由站点管理者按层次结构进行分配;接口ID被用来标识一条链路上的接口,一般情况下(起始不是二进制000)接口ID是按照改进的EUI64格式构建的64位标识符。
参考文献
[1]RFC3769-2004 Requirements for IPv6 Prefix Delegation[s].
[2]RFC3513-2003 Internet Protocol Version 6 ( IPv6) Addr essing Architecture[s].
[3]RFC3177-2001 IAB/IESG Recommendations on IPv6 Address Allocations to Sites[s].
[4]RFC3531-2003 A Flexible Method for Managing the Assig nment of Bits of an IPv6 Address Block[s].
[5]RFC3587-2003 IPv6 Global Unicast Address Formal[s].(end)