Delay-Aware Multipath Parallel SFC Orchestration

Abstract

With the development of network functions virtualization and software defined networking, the service function chain (SFC) orchestration issue is a big challenge for high reliability and low latency services. At present, many studies propose solutions in terms of physical node mapping or link mapping for SFC. In this paper, we consider parallel transmission by dividing the SFC request flow into multiple subflows. To solve the problem of orchestrating parallel SFC under the premise of being able to meet the delay requirements of delay-sensitive classes of services, we divide the problem into two parts: virtual network functions mapped to physical servers and virtual links mapped to physical links. In the first part, we find suitable physical nodes for deployment by the simulated annealing algorithm. In the second part, we construct the link mapping problem as a multi-objective optimization problem. We solve this multi-objective optimization problem by quantum genetic algorithm. Finally, the mapping scheme for parallel SFC is generated. We have conducted comparative analyses of the algorithms through simulation experiments. The results show that the method proposed in this paper can effectively improve the orchestration efficiency of parallel SFC. The algorithm we build can not only minimize the resource consumption and routing energy consumption but also meet the delay requirements well. Therefore, this paper has high practical significance in diverse delay-sensitive service applications and provides a solution for future multipath parallel SFC orchestration.

DRL based SFC Orchestration in SDN/NFV Environments subject to Transient Unavailability

“…However, whenever enough server resources are available, A2C brings better acceptance rate. Authors in [22] formulate multi-path parallel SFC orchestration issue for high reliability and low latency services as a multi-objective optimization problem solved using quantum genetic algorithm. They show that the proposed method can effectively improve orchestration efficiency of parallel SFC through both minimizing resource consumption and routing energy consumption while meeting delay requirements.…”

An Improved Adaptive Service Function Chain Mapping Method Based on Deep Reinforcement Learning

“…Most of the research on SFC mapping problems is concentrated in diverse network scenarios such as 5G networks, the Internet of Things, and the Internet [7]. Wei et al [8] proposed a delay-aware multi-path parallel SFC mapping method. This method divides SFCRs into multiple streams for parallel transmission, and solves the mapping optimization problem of multiple SFC streams by a quantum genetic algorithm.…”

“…where {VNFIs i (t)|VNF x = 1} represents the deployment status of type x VNFs deployed at time slot t, {VNFIs i (t − 1)|V NF x = 0} indicates that type x VNFs are non-deployed at time slot t − 1. s(V NF x ) is the cost of this VNFI deployment. Therefore, the total service cost in time slot t is shown in Equation (8).…”

Intelligent Deployment of Delay-Sensitive Service Function Chain Based on Parallelization and Improved Cuckoo Search Algorithm

“…However, these studies are based on the serial structure of SFC. It is possible to achieve delay optimization by preprocessing through parallelization before the deployment of SFC requests [23][24][25][26]. According to research by Sun et al [23], 53.8% of network functions can be parallelized, and 41.5% of parallelized network functions do not result in additional resource costs.…”

“…Cai et al [25] have proposed an adaptive parallel service graph optimization mechanism to reduce excessive resource consumption caused by the unreasonable service graph. Wei et al [26] have divided SFC into multiple substreams for parallel transmission to meet the delay requirements of delaysensitive services and solved optimization problems through genetic algorithms.…”