A Parallel Cloud Service System and Construction Method Based on Performance Requirements

Abstract

The present invention provides a parallel-serialization cloud service system based on performance requirements, including: a computing cluster and a parallelization module; wherein, the computing cluster includes at least one physical node; and the parallelization module is based on preset parameters , the hardware parameters of each of the physical nodes and the preset virtual machine construction instructions to construct at least one virtual machine on at least one of the physical nodes; the present invention also provides a construction of a parallel cloud service system based on performance requirements method, according to the preset virtual machine construction instructions, the virtual machine can be distributed on multiple physical nodes in various ways, so that the performance of the virtual machine can better meet the user's task requirements, and the user can arrange the virtual machine more flexibly and efficiently .

Description

[0001] This application claims the priority of the Chinese patent application filed with the Chinese Patent Office on December 1, 2016, with the application number 201611089011.7, and the invention title is "a performance-based concurrent cloud service system and construction method", all of which The content is incorporated in this application by reference. Technical field [0002] The present invention relates to the field of cloud computing, in particular to a parallelized cloud service system and construction method based on performance requirements. Background technique [0003] Cloud computing is another huge change after the big transformation from mainframe computers to client-servers in the 1980s. It is distributed computing (Distributed Computing), parallel computing (Parallel Computing), utility computing (Utility Computing), network storage (Network Storage) Storage Technologies, Virtualization, Load Balance, Hot Backup Redundancy (High Available) and other traditional...

AI Technical Summary

Problems solved by technology

[0005] Most of the existing cloud service platforms use the form of creating multiple virtual machines on one physical machine. Users need to set the size of the virtual machine when applying for a virtual machine, and the cloud service provider creates the virtual machine on the physical machine according to the user's needs. In this way, firstly, the user does not know the specific performance of the physical machine, and does not have a good control over the performance of the virtual machine; secondly, the cloud service provider does not know the specific task requirements of the user, and the physical machine allocated for it In terms of performance, it may not be able to meet the user's task requirements well; moreover, in this way, the performance of the virtual machine is limited by the single physical machine itself. It may cause the virtual machine to be overloaded and cause downtime, and may even cause damage to the performance of the physical machine due to long-term overloaded operation

Images