Deep Reinforcement Learning Based VNF Management in Geo-distributed Edge Computing
- 1 July 2019
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS)
- p. 934-943
- https://doi.org/10.1109/icdcs.2019.00097
Abstract
Edge computing is an effective approach for resource provisioning at the network edge to host virtualized network functions (VNF). Considering the cost diversity in edge computing, from the perspective of service providers, it is significant to orchestrate the VNFs and schedule the traffic flows for network utility maximization (NUM) as it implies maximal revenue. However, traditional model-based optimization methods usually follow some assumptions and impose certain limitations. In this paper, inspired by the success of deep reinforcement learning in solving complicated control problems, we propose a deep deterministic policy gradients (DDPG) based algorithm. We first formulate the NUM problem with the consideration of end-to-end delays and various operation costs into a non-convex optimization problem and prove it to be NP-hard. We then redesign the exploration method and invent a dual replay buffer structure to customize the DDPG. Meanwhile, we also apply our formulation to guide our replay buffer update. Through extensive trace-driven experiments, we show the high efficiency of our customized DDPG based algorithm as it significantly outperforms both model-based methods and traditional non-customized DDPG based algorithm.Keywords
This publication has 18 references indexed in Scilit:
- Approximation and Online Algorithms for NFV-Enabled Multicasting in SDNsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- A Scalable and Distributed Approach for NFV Service Chain Cost MinimizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Reinforcement Learning for Energy Harvesting Decode-and-Forward Two-Hop CommunicationsIEEE Transactions on Green Communications and Networking, 2017
- Proactive VNF provisioning with multi-timescale cloud resources: Fusing online learning and online optimizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Provably efficient algorithms for joint placement and allocation of virtual network functionsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Resource Management with Deep Reinforcement LearningPublished by Association for Computing Machinery (ACM) ,2016
- Inapproximability of the Multilevel Uncapacitated Facility Location ProblemACM Transactions on Algorithms, 2016
- Network functions virtualization with soft real-time guaranteesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- Deploying chains of virtual network functions: On the relation between link and server usagePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- Mastering the game of Go with deep neural networks and tree searchNature, 2016