Integration method, device and system for computational resources

Abstract

The embodiment of the invention relates to an integration method, device and system for computational resources. A data center comprises a consistency system composed of a first number of computational nodes and a non-consistency system composed of a second number of computational nodes. The computational nodes are used for providing computation resources. The method comprises following steps: obtaining information on streaming figures of the consistency system; determining strategy information of integration of computational resources based on streaming figures of the consistency system, wherein strategy information comprises node labels and migration paths of computational nodes for system migration; and integrating computational resources based on node labels and migration paths. In the embodiment provided by the integration method, device and system for computational resources, hardware components are not needed to be changed when concrete application scenes change in a data center. Due to system migration, mutual dynamic migration between the consistency system and the non-consistency system are achieved due to system migration. Therefore, system-wide optimal performance and optimal response time can be achieved.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a computing resource integration method, device and system. Background technique [0002] With the continuous development of computer technology, a data center including a plurality of computing nodes appears. The computing nodes are used to provide computing resources. The data center integrates computing resources to collaboratively process computing tasks. [0003] In the prior art, there are various methods for integrating computing resources, but they all adopt a fixed method for integration. refer to figure 1 The schematic diagram of the data center architecture in the prior art is shown, and the computing node is referred to as a node (Node) for convenience of description. The data center includes a 4-node minicomputer and 16 single-node hosts, which together form a computing cluster (Cluster), wherein the 4-node minicomputer is a high-speed cache non-uniform memory ac...

AI Technical Summary

Problems solved by technology

[0005] It can be seen from the above that, with the integration method of computing resources in the prior art, when the specific application scenarios of the data center change, the above-mentioned first method needs to change the hardware composition of the data center; the above-mentioned second method, due to IB or Ethernet The delay of the network makes the response time of the computing task too long

Images