2021/06/28 by Joseph Billingsley, Ke Li, Billingsley, Joseph +5 · 1 citation
Computer Science · Engineering · #Advanced Optical Network Technologies #Cloud Computing and Resource Management #FOS: Computer and information sciences #Networking and Internet Architecture (cs.NI) #Software-Defined Networks and 5G
paper · pdf · doi:10.48550/arxiv.2106.14727
openalex publication_date 2021/06/28 · openalex created_date 2022/09/16 · openalex updated_date 2026/07/28
Data centers are critical to the commercial and social activities of modern\nsociety but are also major electricity consumers. To minimize their\nenvironmental impact, it is imperative to make data centers more energy\nefficient while maintaining a high quality of service (QoS). Bearing this\nconsideration in mind, we develop an analytical model using queueing theory for\nevaluating the QoS of a data center. Furthermore, based on this model, we\ndevelop a domain-specific evolutionary optimization framework featuring a\ntailored solution representation and a constraint-aware initialization operator\nfor finding the optimal placement of virtual network functions in a data center\nthat optimizes multiple conflicting objectives with regard to energy\nconsumption and QoS. In particular, our framework is applicable to any existing\nevolutionary multi-objective optimization algorithm in a plug-in manner.\nExtensive experiments validate the efficiency and accuracy of our QoS model as\nwell as the effectiveness of our tailored algorithms for virtual network\nfunction placement problems at various scales.\n