The dramatic increasing of data and demands for computing capabilities may result in excessive use of resources in cloud data centers, which not only causes the raising of energy consumption, but also leads to the violation of Service Level Agreement (SLA). Dynamic consolidation of virtual machines (VMs) is proven to be an efficient way to tackle this issue. In this paper, we present an Adaptive Deep Reinforcement Learning (DRL)-based Virtual Machine Consolidation (ADVMC) framework for energy-efficient cloud data centers. ADVMC has two phases. In the first phase, Influence Coefficient is introduced to measure the impact of a VM on producing host overload, and a dynamic Influence Coefficient-based VM selection algorithm (ICVMS) is proposed to preferentially choose those VMs with the greatest impact for migration in order to remove the excessive workloads of the overloaded host quickly and accurately. In the second phase, a Prediction Aware DRL-based VM placement method (PADRL) is further proposed to automatically find suitable hosts for VMs to be migrated, in which a state prediction network is designed based on LSTM to provide DRL-based model more reasonable environment states so as to accelerate the convergence of DRL. Simulation experiments on the real-world workload provided by Google Cluster Trace have shown that our ADVMC approach can largely cut down system energy consumption and reduce SLA violation of users as compared to many other VM consolidation policies.

Intelligent task scheduling solutions are highly demanded in the operation of complex cloud data centers so that resources can be utilized in an energy-efficient way while still ensuring various requirements of users. However, the energy problem of task scheduling in cloud environment becomes more challenging with the ever-increasing number of users as well as the constant and unpredictable change of workloads. In this research, we propose an Adaptive Deep Reinforcement Learning-based (ADRL) task scheduling framework for energy-efficient cloud computing. We first present a Change Detection algorithm to detect whether the workload has changed greatly. On this basis, we built an Automatic Generation network to adjust the discount factor of Deep Reinforcement Learning (DRL) dynamically according to the changing workload, which enables faster and more accurate learning. We finally introduce the adaptive DRL to learn the optimal policy of dispatching arriving user requests with the reward aiming to minimize task response time and maximize resource utilization. Simulated experiments have confirmed that the proposed scheduling scheme performs well on accelerating learning convergence and promoting allocation accuracy, thus it is very effective in reducing the average response time of tasks and increasing the CPU utilization rate of resources, which eventually makes the cloud system more energy efficient.

For the wireless sensor networks (WSNs) heterogeneous node deployment optimization problem with obstacles in the monitoring area, two new flower pollination algorithms (FPA) are proposed to deploy the network. Firstly, an improved flower pollination algorithm (IFPA) is proposed based on FPA, aiming at the shortcomings of the convergence speed is slow and the precision is not high enough of FPA. The nonlinear convergence factor is designed to correct the scaling factor of FPA, the Tent chaotic map effectively maintains the diversity of the population in the late iteration, and a greedy crossover strategy is designed to assist the remaining individual search with better individuals. Secondly, based on FPA, a non-dominated sorting multi-objective flower pollination algorithm (NSMOFPA) is proposed. The external archive strategy and leader strategy are introduced, to solve the global pollination problem. The proposed crowding degree method and the introduced elite strategy effectively maintain the diversity of the population. Then, IFPA is applied to WSN deployment aiming at optimizing coverage rate, simulation experiments show that IFPA can obtain a higher coverage rate with shorter iterations, which can save network deployment costs. Finally, applying NSMOFPA to the WSN deployment with optimization objectives for coverage rate, node radiation overflow rate and energy consumption rate. The experimental results verify that NSMOFPA has a good optimization effect and can provide a better solution for WSN deployment.

Aiming at the difficulty of deploying wireless sensor networks (WSNs) on three-dimensional (3D) surfaces, based on the grey wolf optimizer (GWO), an enhanced version of the grey wolf optimizer is proposed for deploying WSNs on 3D surfaces, namely the enhanced grey wolf optimizer (EGWO), which is characterized by enhanced exploitation and exploration ability of the algorithm. The novelty of EGWO is that the grey wolf population is divided into two parts, one part is responsible for the outer-layer encircle and the other is responsible for the inner-layer encircle, and the introduction of Tent mapping. The purpose of this is to enhance the exploitation and exploration ability of the algorithm respectively, so as to improve the convergence and optimization precision of the algorithm. In addition, in terms of WSN deployment in 3D surfaces, this paper improves the means of determining the perceived blind zone. Meanwhile, a novel method to calculate the WSNs coverage area of simple and complex 3D surfaces is presented by combining the grid and integral of the 3D surfaces. The EGWO is favorably compared with the GWO and three existing variants of the grey wolf optimizer when testing on 12 well-known benchmark functions. The simulation experiment results show that compared with the existing algorithms, EGWO can provide a very competitive search result in terms of optimization precision and convergence performance. Finally, this paper applies EGWO to the 3D surface deployment of WSN. Simulations show that compared with the other three deployment algorithms, EGWO can improve the network coverage of WSN, which can save network deployment costs. In addition, the probability of network connectivity deployed by EGWO is higher, that is, EGWO can provide a better deployment solution.

Managing scientific applications in the Cloud poses many challenges in terms of workflow scheduling, especially in handling multi-objective workflow scheduling under quality of service (QoS) constraints. However, most studies address the workflow scheduling problem on the premise of the unchanged environment, without considering the high dynamics of the Cloud. In this paper, we model the constrained workflow scheduling in a dynamic Cloud environment as a dynamic multi-objective optimization problem with preferences, and propose a transfer learning based multi-objective evolutionary algorithm (TL-MOEA) to tackle the workflow scheduling problem of dynamic nature. Specifically, an elite-led transfer learning strategy is proposed to explore effective parameter adaptation for the MOEA by transferring helpful knowledge from elite solutions in the past environment to accelerate the optimization process. In addition, a multi-space diversity learning strategy is developed to maintain the diversity of the population. To satisfy various QoS constraints of workflow scheduling, a preference-based selection strategy is further designed to enable promising solutions for each iteration. Extensive experiments on five well-known scientific workflows demonstrate that TL-MOEA can achieve highly competitive performance compared to several state-of-art algorithms, and can obtain triple win solutions with optimization objectives of minimizing makespan, cost and energy consumption for dynamic workflow scheduling with user-defined constraints.