Container Migration Mechanism for Load Balancing in Edge Network Under Power Internet of Things

Abstract

As a novel computing technology closer to business ends, edge computing has become an effective solution for delay sensitive business of power Internet of Things (IoT). However, the uneven spatial and temporal distribution of business requests in edge network leads to a significant difference in business busyness between edge nodes. Due to the natural lightweight and portability, container migration has become a critical technology for load balancing, thereby optimizing the resource utilization of edge nodes. To this end, this paper proposes a container migration-based decision-making (CMDM) mechanism in power IoT. First, a load differentiation matrix model between edge nodes is constructed to determine the timing of container migration. Then, a container migration model of load balancing joint migration cost (LBJC) is established to minimize the impact of container migration while balancing the load of edge network. Finally, the migration priority of containers is determined from the perspective of resource correlation and business relevance, and by introducing a pseudo-random ratio rule and combining the local pheromone evaporation with global pheromone update at the same time, a migration algorithm based on modified Ant Colony System (MACS) is designed to utilize the LBJC model and thus guiding the choice of possible migration actions. The simulation results show that compared with genetic algorithm (GA) and Space Aware Best Fit Decreasing (SABFD) algorithm, the comprehensive performance of CMDM in load balancing joint migration cost is improved by 7.3% and 12.5% respectively. INDEX TERMS container migration, load balancing, migration cost, Edge computing, power Internet of things.

Container Placement and Migration in Edge Computing: Concept and Scheduling Models

“…In containers, the memory allocated to running applications is run on top of a hosts operating system's kernel [38]. A container can be described as a live application environment that has occupied a portion of computing resources of a host machine [39].…”

“…The container placement and migration problem in edge computing can be modelled using optimization modelling, as it consists of a set of decision variables, decision constraints, objective functions and model assumptions [84][85][86]. In the container placement problem in edge computing, the objective function relates to the goals such as minimizing latency [84], minimizing overall cost of providing edge services [12,39], minimizing total energy used by the server [86] and minimizing load imbalance [39]. Satisfying SLA is a maximization objective [10,33].…”

“…Satisfying SLA is a maximization objective [10,33]. Many of the problems tend towards a joint optimization problem, see examples in [14,39,71,[84][85][86]. This addressed the tradeoffs that may exist while attempting to fulfil one objective over another.…”

“…The decision variables are the scheduling decisions to be taken, such as which computing tasks to accept [10,87], what containers will the task be assigned to [87], what edge nodes will the containers be deployed [33] or migrated [39]. Constraints in optimization problems can be described as finite resources that the decision maker has to consider when taking the decision.…”

“…Another heuristic-based algorithm is Ant Colony. It can be used for finding the optimal paths through directed graph networks [39]. The Ant Colony algorithm is non-iterative, rather it uses random sampling probability to take decisions.…”