Data center physical resource floating method

Abstract

The present invention provides a data center physical resource floating method. The method includes the following steps that: one network card network port of all physical hosts in each computing cluster, adopted as a mirroring network, is connected with a mirroring switch, and another network, adopted as a host network, is connected with a host switch; a physical host is selected from one computing cluster, mirroring network startup is set for the physical host, and the operating system of the physical host is cloned into mirroring templates through the mirroring network; the switch configuration file of a network required by the computing cluster is saved to the host switch so as be made into a network configuration template; and the mirroring templates are restored in batches to an idlephysical host in another computing cluster through the mirroring network; and the network configuration template is loaded to the host switch of the computing cluster to which the idle physical hostbelongs. With the data center physical resource floating method of the invention adopted, physical resources can be flexibly flexed in batches between the computing clusters, and therefore, a physicalresource floating effect can be achieved, and a resource utilization rate can be improved.

Description

technical field [0001] The invention relates to the field of IT, in particular to a method for floating data center physical resources. Background technique [0002] In recent years, the vigorous development of cloud computing technology has brought great changes to the entire IT industry. As the infrastructure of cloud computing and the innovation platform of next-generation network technology, the research on data center resources has become a hot spot in the academic and industrial circles in recent years. In the new application mode, there are multiple systems and cloud computing clusters in the data center. The traditional hierarchical data center network topology construction method is rigid, and there are many deficiencies in scale, scalability, utilization and cost, and cannot support multiple Rapid elastic scaling between computing clusters has been difficult to meet the growing demand for high performance, high cost performance, and high utilization under the new ...

AI Technical Summary

Problems solved by technology

In the new application mode, there are multiple systems and cloud computing clusters in the data center. The traditional hierarchical data center network topology construction method is solidified, and there are many deficiencies in scale, scalability, utilization and cost, and cannot support multiple Rapid elastic scaling between computing clusters has been difficult to meet the growing demand for high performance, high cost performance, and high utilization under the new situation

Images