Search

Energy efficiency is a critical issue in the management and operation of cloud data centers, which form the backbone of cloud computing. Virtual machine (VM) placement has a significant impact on energy-efficiency improvement for virtualized data centers. Among various methods to solve the VM-placement problem, the genetic algorithm (GA) has been well accepted for the quality of its solution. However, GA is also computationally demanding, particularly in the computation of its fitness function. This limits its application in large-scale systems or specific scenarios where a fast VM-placement solution of good quality is required. Our analysis in this paper reveals that the execution time of the standard GA is mostly consumed in the computation of its fitness function. Therefore, this paper designs a data structure extended from a previous study to reduce the complexity of the fitness computation from quadratic to linear one with respect to the input size of the VM-placement problem. Incorporating with this data structure, an alternative fitness function is proposed to reduce the number of instructions significantly, further improving the execution-time performance of GA. Experimental studies show that our approach achieves 11 times acceleration of GA computation for energy-efficient VM placement in large-scale data centers with about 1500 physical machines in size. [ABSTRACT FROM AUTHOR]

Energy efficiency is a critical issue in the management of data centers, which form the backbone of cloud computing. Virtual resource management has a significant impact on improving the energy efficiency of data centers. Despite the progress in this area, virtual resource management has been considered mainly at two separate levels: application assignment and virtual machine placement. It has not been well-investigated in a unified framework for both levels, limiting further improvement in the energy efficiency of data centers. To address this issue, this paper proposes the virtual resource management problem for energy efficiency as a constrained optimization problem. Then, this paper simplifies the problem through profile-guided task classification and problem decomposition for complexity reduction and improved energy efficiency. After that, a three-phase framework and algorithms are presented for profiling and profile updating, task classification and application assignment, and successive virtual machine placement. Experimental studies show energy savings of 8–12% by the three-phase framework compared to the existing technique. [ABSTRACT FROM AUTHOR]

Energy efficiency is a critical issue in the management and operation of data centers, which form the backbone of cloud computing. Virtual machine placement has a significant impact on the energy efficiency of virtualized data centers. Among various methods to solve the virtual-machine placement problem, the genetic algorithm has been well accepted for its quality of solution. However, it is computationally demanding largely due to the complex form of fitness function, limiting its applications in data centers. To enhance the computational efficiency of the genetic algorithm while maintaining its quality of solution, a progressive-fidelity approach is developed in this paper for genetic-algorithm computation. It starts with a low-fidelity genetic algorithm with a simple fitness function. Then, for solution refinement, it switches to a medium-fidelity genetic algorithm with a more complicated fitness function. Finally, it progresses to the fine tuning of solution through a high-fidelity genetic algorithm with the energy consumption of data centers as fitness function. Heuristics are presented for the adaptive switching of genetic-algorithm computation from low fidelity to medium fidelity and finally to high fidelity. Experiments show that compared with the standard genetic algorithm of high fidelity, our progressive-fidelity approach of genetic algorithm computation is 50% faster for large-scale data centers while maintaining similar quality of solution in terms of the energy consumption of data centers. • A progressive-fidelity approach developed for speeding up Genetic Algorithm. • 3 different fitness functions included for a quick and fine solution. • Heuristics are presented for the adaptive switching in our approach. [ABSTRACT FROM AUTHOR]

[Display omitted] • TA C appr. is proposed as a novel objective function for DWC parameter optimization. • Optimal energy control structure can be achieved by manipulating liquid split ratio. • MPC-PI hybrid structure balances both control performance and safety performance. • Simulated annealing algorithm is employed to optimize the weights of MPC controller. Polysilicon quality and energy consumption are directly affected by the purification process of trichlorosilane. In this work, the dividing wall column (DWC) as a promising energy-saving technology is utilized for trichlorosilane purification, with the design optimization carried out using steady-state simulation. Compared with conventional distillation process, DWC can reduce total annual cost (TAC), CO 2 emissions and exergy loss (El) by 35.81%, 53.56% and 54.10% on average, respectively. Established on the characteristics of superfractionator, four control structures are proposed in this work, two of which are multi-loop proportional-integral (PI) control schemes including condenser duty (C), distillate flow rate (D), bottom flow rate (B)/reflux flow rate (R), side flow rate (S), reboiler duty (V) control structure and optimal energy control structure. The other two are model predictive control (MPC) schemes including standard MPC structure and MPC-PI hybrid structure. The weights of MPC controller are optimized using the simulated annealing (SA) algorithm. Dynamic simulation results demonstrate that MPC schemes achieve improved closed-loop performance in terms of minor overshoot, shorter transition time and reduced oscillation than PI control schemes. The integral absolute error (IAE) is introduced to further quantitatively evaluate MPC schemes. The simulation results demonstrate that the standard MPC structure achieves the best control performance and the MPC-PI hybrid structure enhances process safety while maintaining the desired control performance. [ABSTRACT FROM AUTHOR]

Energy efficiency is a critical issue in data centre management, which is the foundation for cloud computing. The VM placement has a considerable impact on a data centre's energy efficiency and resource utilisation. The assignment of VMs to PMs is an NP-hard problem without an easy way to find an optimal solution, particularly in large-scale data centres. In this study, the VM placement problem is formulated as a constrained optimisation problem. The Genetic Algorithm (GA) is a suitable method for solving this problem in terms of the quality of the solution. However, GA is time-consuming to obtain an optimal solution in the large scale optimisation problem. Therefore, this paper focuses on accelerated GA for energy-efficient VM placement. As the most time-consuming element of the GA is the calculation of its fitness function, this paper simplifies this calculation through a new fitness function in GA. Simulation results of small-, medium-, and large-scale data centres demonstrate that our accelerated GA is faster than the standard GA and gives better quality of solution than the First Fit Decreasing (FFD) algorithm, respectively. The findings of our GA with the new fitness function reveal an 8% energy saving for our GA compared to FFD and a 66% reduction in our GA execution time compared to the standard GA with standard energy formula as a fitness function. The number of generations in our GA is reduced by about 50% in comparison with the standard GA. Moreover, we started with 3000 PMs in the large-scale dataset, and only 1086 PMs were actually used after running our GA. Therefore, we may switch off far more PMs for energy savings from our GA results than those from the standard GA. • Virtual machine placement problem formulated as a constrained optimisation problem. •It focuses on accelerated genetic algorithm for virtual machine placement. •Our algorithm can minimise residual resources using Taylor extension. •Our solution quality is better than FFD and standard genetic algorithm. •It reduces execution time and energy in large scale data centres. [ABSTRACT FROM AUTHOR]