Research On Service-oriented And Energy-efficient Scheduling Algorithms In Data Center

In cloud computing,it is crucial to maintain service continuity,while power outage is one of the most common and serious threats.To improve the resilience of cloud against power outage,a service provider usually deploys emergency energy supply in a data center.However,due to limited emergency supply,cloud service providers need to provide feasible solutions and design energy-efficient scheduling algorithms to ensure continuity of cloud service.However,we notice that current works lack theoretical research on maintaining service continuity under power outage.Based on this,we study the recovery/continuity problems of cloud service under two different scenarios.We first study the scenario of a homogeneous network service where the workload changes dynamically in real time.To deal with power outage,we introduce a hybrid cloud scheduling model by introducing a public cloud to processa part of the workload.When a power outage occurs,the cloud service provider needs to determine the workload distribution and dynamically adjust computing resources according to the currently arrived workload to reduce the cost of renting a public cloud.We study the online cloud service recovery problem under power outage.By using Lyapunov optimization technology,the long-term online optimization problem is transformed into a general optimization problem.We present a power-aware online control algorithm(POCA)that achieves an [O(V),O(1/V)] performance trade-off between power consumption and cost.A large value of V will result in increased energy consumption but a reduced disaster recovery cost.Simulation experiments verify the performance of the proposed algorithm.We further study the scenario where multiple heterogeneous services are deployed in general cloud data centers.When a power outage happens,the cloud service provider needs to determine VM selection strategy and VM consolidation strategy in order to use emergency power supply to minimize its loss(or maximize its profit)while ensuring the requirement of the service continuity.We study the cloud service continuity problem in this scenario.We first formulate an optimization problem that aims to maximize the total profit subject to the limited energy constraint.After showing the hardness of the problem,we focus on the design of approximation algorithms for solving the problem,where we consider two practical cases.In the first one with sufficient number of servers for re-provisioning,we develop a constant approximation algorithm of which the worst-case performance approaches the optimal solution within a constant factor(? 4.5-6.4).In the second one,we consider the general case with limited number of servers,and we develop an approximation algorithm with an approximation ratio of around 5.7-8.By combining these two algorithms together,we can achieve both good worstcase performance and average performance.Simulation results demonstrate the efficiency in terms of maximizing the service continuity profit of the proposed algorithms.