An Algorithm To Solve The Block Bordered Diagonal Form In Data Center

With the development of large-scale regional interconnected power grids,power flow computation, transient analysis, reactive power optimization in power system become increasingly complex, which put forward higher requirements on the power system fast processing of online complex calculation. Block Bordered Diagonal Form(BBDF)as the power system's basic arithmetic unit, is widely used in large-scale power network computing, enhancing its operational efficiency will greatly accelerate system analysis and energy efficiency and meet the requirements of large-scale real-time online calculation. However, most of traditional solving methods adopted single server or simple scheduling multi- servers, with long running time and unsatisfactory result.Cloud computing, a novel form of data storage on the Internet, increases the possibility to implement a tool for analyzing large-scale of online calculation in power systems and parallel computing in complex grids, in the processing of cloud computing, the data center energy consumption is equally important with speed up. In this paper, we propose a new optimized method for the mapping of BBDF and decomposition-coordination algorithms for cloud computing data centers. Based on the computational complexity of coupling between tasks, the decomposition-coordination algorithm is split to perform different tasks, judging on the amount of calculation. A binding placement algorithm is also presented as a method to map the tasks into virtual machines(VM). A new energy-efficient Bin-Packing model is also built for the final mapping step, which is the process of transferring the data from the VMs to the data centers. This will be performed while ensuring that the constraints of CPU and memory utilization rate are in check. IEEE 118 node grid models, as well as two large-scale power systems, which utilizes 538 and 1133 nodes systems, are calculated though t he CloudSim platform. The results suggest that the method of calculating application data centers using virtual machine technology is more effective than the use of traditional parallel computing methods on computing time and energy consumption, besides, t he binding placement algorithm is more effective than the sequence placement algorithm; further more binding-decrease best fit algorithm is the best method of them. In the IEEE 118 nodes system, the total processing time and energy have reduces 42.44% and 48.04%, in the 538 nodes system, the total processing time and energy have reduces 69.08% and 75.27%, and in the 1133 nodes system, the total processing time and energy have reduces 75.79% and 80.85%. With the increasing scale computing, data center computing ability is more and more obvious.