Optimizing Service Chain ID generation for flow rule compression

Abstract

Service Chaining provides opportunities for network and service providers to implement their services and policies with finer granularity of an individual user or an application. However, the increasing number of Service Chains and middleboxes will introduce a larger number of flow rules and more consumption of Ternary Content Addressable Memory (TCAM), whose capacity is limited due to high cost and power consumption. This paper proposes to compress the flow rules for service chaining by optimizing the generation of Service Chain IDs that are widely used in packet tagging techniques for the Service Chaining. Our solution 1) makes service chain IDs aggregatable based on Common Forwarding Actions (CFAs) among the service chains, and 2) reduces the number of flow rules at each SDN switch to execute a larger number of forwarding actions for service chaining. The evaluation results showed that the proposed algorithm can reduce up to 76% of the flow rules using the randomly generated networks and service chains. Because the generation of Service Chain ID does not interfere the other flow rule compression techniques, our algorithm can also be used as a plug-in to the other Service Chaining mechanisms to optimize their ID generation.