Service-centric traffic engineering and cache orchestration in an ICN-enabled network

Abstract

Large service providers (SP) who own their computing, storage, and networking resources have the potential to jointly manage their resources to achieve better quality of service and lower cost. Information-centric networking (ICN) exploits in-network caching resource, by unique identification of content objects. Orchestrating cache and networking resources in a coordinated way is the key to the success of ICN. Traffic engineering (TE), which allocates networking resources to services, face new challenges in the presence of geographically- replicated services and in-network cache in ICN. Some previous works [1] [2] [3] have addressed this by jointly optimizing content placement (allocating storage resources) and TE. However, jointly managing storage and networking resources at the content level is not scalable because of the huge number of contents and nontrivial amount of overhead to distribute content information and optimize content caching and routing at small time scale. In this study, we solve the TE and cache orchestration (CO) problems in an ICN-enabled network at the service level, in which no specific information about individual contents is needed. So, the proposed optimization is both scalable and easily implementable. Specifically, we formulate two simple but effective linear programs to solve the TE problem at the service level for scenarios with and without caching. We also propose a general network-flow model to solve the service-level CO problem. Simulation results show that our proposed TE and CO schemes can significantly reduce network cost and increase cache hit ratio, with the constraints of satisfying service-level agreements (SLA).