作者：
Wei, WeiLi, HaoyiZhang, QinghuiHenan Univ Technol
KeyLab Grain Informat Proc & Control Minist Educ Zhengzhou 450001 Peoples R China Henan Univ Technol
Henan Key Lab Grain Storage Informat Intelligent P Zhengzhou 450001 Peoples R China Henan Univ Technol
Henan Grain Big Data Anal & Applicat Engn Res Ctr Zhengzhou 450001 Peoples R China
With the development of cloud technology, hierarchical and distributed cross-cloud architecture is gradually replacing traditional centralized architecture, for example, used in edge (or fog) computing. Due to the flu...
详细信息
With the development of cloud technology, hierarchical and distributed cross-cloud architecture is gradually replacing traditional centralized architecture, for example, used in edge (or fog) computing. Due to the fluctuation of resource requirements, if a node does not have sufficient resources to process requests, the same or higher-level nodes can share their resources by offloading or redirecting requests to themselves, at the possible cost of reduced service quality. However, it is difficult to effectively optimize the sharing effect based on mean requirements. We formulate the multilevel problem with horizontal and vertical resource sharing using stochastic models, identify the optimal structures with embedded subproblems, and obtain the approximation solution in an efficient dynamic programming manner. In the problem setting with a wide range of different parameters, the proposed algorithm can outperform existing mean and heuristic algorithms in all scenarios to improve the total satisfied requirements by up to 26%, and can be hundreds of times faster than these heuristic algorithms.
To support distributed upper-level services such as those in edge or fog computing, a hierarchical and distributed multicloud architecture is introduced to replace the traditional centralized cloud architecture. The i...
详细信息
To support distributed upper-level services such as those in edge or fog computing, a hierarchical and distributed multicloud architecture is introduced to replace the traditional centralized cloud architecture. The intra-time slot demand fluctuation with corresponding multidimensional resource sharing are neglected for efficiency reasons in the existing algorithms. By considering them, the service provider has the potential to obtain better solution, but also makes the scheduling problem a highly complicated stochastic optimization one. To effectively and efficiently solve the problem containing a nonlinear objective and constraints with heterogeneous nonlinear pricing functions, the general stochastic scheduling problem is formulated and its structure is exploited to obtain the base solution quickly, which is further applied with solution closure for fast convergence in the following configurable metaheuristic searching phrase. The experiments with many different problem settings show its superiority over existing algorithms with more than 30% revenue improvement. Therefore, the proposed algorithm can be used as a good supplement to the existing methods.(c) 2022 Elsevier B.V. All rights reserved.
暂无评论