Carrier-grade availability-aware mapping of Service Function Chains with on-site backups

Abstract

Network Function Virtualization (NFV) is a promising technique to greatly improve the effectiveness and flexibility of network services through a process named Service Function Chain (SFC) mapping, with which network functions (NFs) are deployed over virtualized and shared platforms in data centers. NFV typically requires a higher availability at the carrier grade than conventional cloud-based IT services, provided by native IaaS mechanism, for example. To achieve the high availability, each VNFs in an SFC can be provisioned with sufficient on-site backups. However, having too many backups may greatly decrease the resource utilization. Therefore, an open challenge is to find an effective method to allocate backup resource in order to maximize the number of SFC requests that can be served while meeting their heterogeneous availability requirements. To address this challenge, we first study how to allocate a minimum amount of backup resource for a single SFC request as an integer nonlinear program and provide an optimal solution. Based on the solution, we then propose an online heuristic algorithm for mapping multiple SFC requests with the objective of maximizing the number of SFC requests that can be served. Last but not least, we introduce a novel backup pooling mechanism to further improve the efficiency of backup resource usage. Through simulations, we show that our proposed algorithm can significantly reduce resource consumption due to backups and increase the number of co-existing SFC requests that can be served.