An efficient parallel IP lookup technique using CREW based multiprocessor organization

Abstract

IP address lookup is a multifaceted problem because of increasing routing table sizes, increased traffic, higher speed links, and the migration to 128 bit IPv6 addresses. Routers bring into play the packet's destination address to determine for each packet the next hop. IP address lookup is complicated because it requires a longest matching prefix search. We propose a CREW based multiprocessor organization lookup (CMOL) that uses contemporaneous sorting and searching. This technique further uses N+1-ary search to speedup the address lookup operations. Since multiple processors are used, the prefixes to be compared are reduced for each processor. The number of comparisons for each processor is reduced by a factor of n/N. The time complexity of searching and sorting has been reduced to O(logN+1(N+1)). The use of controlled prefix expansion further reduces the storage space