Research Of Sensor Resource Scheduling Optimization Technology In Grid

Grid is a new distributed computing technology, it makes full use of the existing hardware and software resources in the Internet to support data, storage, information and resource sharing to eliminate information silos, to solve large-scale application of collaborative and achieve high performance at low cost. The grid resources have distributed, dynamic, heterogeneous and autonomous characteristics, so the management of grid resources and resource scheduling become a key issue of the grid research. Scheduling the resources in a reasonable way is important for the performance of grid. Traditional resource-scheduling algorithm can solve the problem effectively, while with the wireless sensor network developing rapidly and sensor nodes being used as grid resources widely in grid system, the problem of energy consumption and QoS constraints of the task is taking a more and more important place in grid resources scheduling. Therefore, it has become an urgent problem in the field of grid resources scheduling to seek a new algorithm for sensor resources scheduling.This thesis analyses the problem of multiple QoS constraints and energy consumption about the sensor nodes scheduling. To solve the problem, an algorithm about sensor nodes scheduling in grid environment is proposed.The thesis focuses on the following researches and innovations:Firstly, this thesis introduces the features, purpose and function of the grid resources management, then expatiate on the importance of the resources management in the grid resources management. The thesis also analyses several traditional heuristic algorithms in detail and compares them in the time complexity respect.Secondly, this thesis analyses the characteristic of sensor nodes scheduling about multiple QoS constrains, then a sensor nodes scheduling model based on multiple QoS constrains is proposed. In the model, grid QoS is graded according to the influence to resource scheduling, and the thesis establishes a rational priority scheduling scheme. Based on the QoS constraints of the tasks, the energy consumption is also considered in the sensor nodes scheduling, a model is made based on energy consumption and energy balance. The model attaches importance to energy consumption and energy balance of the sensor nodes about resources scheduling to maximize the use time of each sensor node.Thirdly, a sensor nodes scheduling algorithm is proposed based on the multiply QoS constrains and energy optimization. Based on the classification of QoS constrains, a tradeoff function is used to estimate the Makespan consumption and energy consumption in a scheduling. The algorithm selects the lowest cost resource and task according to the tradeoff function to achieve energy optimization and energy balance.Finally, the SimGrid Toolkit is selected for testing the performance of the sensor nodes scheduling algorithm. Several group experiments are designed to compare the new algorithm with other two traditional grid resource scheduling algorithms, which confirms the correctness and feasibility of the algorithms in practice.This thesis is supported by National Natural Science Foundation of China (NSF) under grant (No.60970064),Open Fund of the State Key Laboratory of Software Development Environment (No:SKLSDE-2011KF-01),Program for New Century Excellent Talents in University, China under grant (NCET-08-0806),Municipality science and technology in the Wuhan offends pass item (No:201010621207),and Fok Ying-Tong Education Foundation for Young Teachers in Higher Education Institutions of China (No:121067).