Rule Set Optimization for Packet Pre-processing using Hash based Algorithm

With the advancement in communication technology, gigabit networks are becoming common place. The traditional processors do not have sufficient processing capabilities to handle and process packets arriving at such high speed. So in order to fulfill the need of high processing, application specific processors are tailored to perform dedicated network specific processing. Packet pre-processing is one such application where the prime focus is on increasing the throughput. The aspect of memory usage for this application is neglected by most of the application which may degrade the performance when the arrival rate of packet is too high. The work presented here primarily focuses on pre-processing of rule-set used in packet processing using quadratic probing hashing technique for optimizing memory usage while maintaining fair throughput. The implementation result shows a maximum of 11% memory consuming compared to actual rule sets size and it has been tested on various rule sets like ACL, FW and IPC.