Journal
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
Volume 38, Issue 7, Pages 1490-1504Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSAC.2020.2986867
Keywords
Redundancy; Resilience; Hardware; Software; Middleboxes; Network function virtualization; Network function virtualization; network resilience; redundancy; network reliability; placement and chaining; network diversity
Ask authors/readers for more resources
Achieving network resiliency in terms of availability, reliability and fault tolerance is a central concern for network designers and operators to achieve business continuity and increase productivity. It is particularly challenging in increasingly virtualized network environments where network services are exposed to both hardware (e.g., bare-metal servers, switches, links, etc.) and software (VNF instances) failures. This increased risk of failures can severely deteriorate the quality of the deployed services and even lead to complete service outages. In this context, deploying services in operational networks often exacerbates the availability problem and requires considering availability of hardware and software components both individually and collectively. A key challenge in this perspective is the additional resources needed to achieve partial or full recovery after failures. In this paper, we propose a joint selective diversity and tailored redundancy mechanism to provision resilient services in an NFV framework. Diversity splits a single VNF into a pool of N active instances called replicas while redundancy provides P standby ready-to-use instances called backups. Based on an enhanced N+P model, we propose a placement solution of Service Function Chains (SFC) modeled as a Mixed Integer Linear Program (MILP). The proposed solution is designed to meet a target SFC availability level and, at the same time, to reduce the inherent cost due to diversity (overhead) and redundancy (backup resources). We evaluate the efficiency of the proposed solution through numerically and experimentally. Results demonstrate that our solution, not only, improves service resiliency by avoiding complete service outages but can also overcome network resource fragmentation.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available