Managing data center resources using heuristic search

• Дата: 2018
• Автор: Zharikov, E.V.
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: Інститут програмних систем НАН України 2018
• Теми: virtualization; resource management; cloud computing; heuristic search; UDC 004.94; 004.4; 004.62
• Онлайн доступ: https://pp.isofts.kiev.ua/index.php/ojs1/article/view/307
• Назва журналу: Problems in programming

The features of the cloud data center are analyzed from the point of view of resource management. The two-stage method for consolidating virtual machines based on the use of local beam search algorithm is proposed and investigated with aim to solve the problem of managing the resources of a cloud data center. In this paper, the work of heuristics of the first and second stages of the proposed method is analyzed. The beam search algorithm was developed for solving the data center resource management problem. The data about tasks and physical machines from the Google cluster-usage traces are used to evaluate the proposed method. The proposed method allows to switch to a low-power mode on average 56 percent of physical servers potentially identified for switching to sleep mode based on an upper estimate of the required capacity of resources. Virtual machine consolidation is performed taking into account the limitation of the permissible number of migrations per physical server.Problems in programming 2017; 4: 016-027