2019/06/12 by Thembelihle Dlamini, Dlamini, Thembelihle, Ángel Fernández Gambı́n +1
Computer Science · #Cloud Computing and Resource Management #FOS: Electrical engineering #IoT and Edge/Fog Computing #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1906.05008
openalex publication_date 2019/06/12 · openalex created_date 2022/07/28 · openalex updated_date 2026/07/28
In virtualized computing platforms, energy consumption is related to the\ncomputing-plus-communication processes. However, most of the proposed energy\nconsumption models and energy saving solutions found in literature consider\nonly the active Virtual Machines (VMs), thus the overall operational energy\nexpenditure is usually related to solely the computation process. To address\nthis shortcoming, in this paper we consider a computing-plus-communication\nenergy model, within the Multi-access Edge Computing (MEC) paradigm, and then\nput forward a combination of a traffic engineering- and MEC Location\nService-based online server management algorithm with Energy Harvesting (EH)\ncapabilities, called Automated Resource Controller for Energy-aware Server\n(ARCES), for autoscaling and reconfiguring the computing-plus-communication\nresources. The main goal is to minimize the overall energy consumption, under\nhard per-task delay constraints (i.e., Quality of Service (QoS)). ARCES jointly\nperforms (i) a short-term server demand and harvested solar energy forecasting,\n(ii) VM soft-scaling, workload and processing rate allocation and lastly, (iii)\nswitching on/off of transmission drivers (i.e., fast tunable lasers) coupled\nwith the location-aware traffic scheduling. Our numerical results reveal that\nARCES achieves on average energy savings of 69%, and an energy consumption\nranging from 31%-45%and from 21%-25% at different values of per-VM\nreconfiguration cost, with respect to the case where no energy management is\napplied.\n