TCAM-PC: Space-efficient TCAM-based packet classification with packet-forwarding-rate constraints

摘要

Packet classification has been the core component of current Internet architecture, and ternary content addressable memory (TCAM) has become the de facto standard for performing high-speed packet classification. Unfortunately, TCAM has limitations such as small capacity and high power consumption. Therefore, how to improve memory usage and decrease power consumption are crucial to TCAM-based technologies. In this paper, we propose a space-efficient TCAM-based packet classification scheme called TCAM-PC, which decreases TCAM memory consumption by compressing the space for the same information parts embedded in different rules, and guarantees a predetermined packet forwarding rate based on the cooperation between the global mask register and block mask register in a TCAM. Performance analysis shows TCAM-PC can achieve smaller memory requirement and higher memory utilization while guaranteeing a high-speed packet forwarding rate. To the best of our knowledge, this is the first attempt to adaptively compress the TCAM space while guaranteeing a high packet forwarding rate.