Cloud computing implementation method and system

Abstract

The invention discloses a cloud computing implementation method. By multilayer load balancing architecture design, a plurality of virtual machines with totally same functions are deployed in an application layer, a middle layer and a database layer, and fault monitoring modules are respectively deployed on the virtual machines; when one virtual machine has a file system read-only fault and the like, the fault monitoring module automatically turns off the virtual machine to realize automatic isolation of a virtual machine node, and reports the fault information to a management center; the management center migrates the access request of the virtual machine having the fault to the other corresponding virtual machine; and the fault or the deletion of any main virtual machine does not cause pause on the current service system, so that the fault of a single main virtual machine can be prevented from affecting the whole cloud computing platform, the smoothness of services is ensured, and the user experience and the performance of the whole system are improved. The invention also discloses a cloud computing system.

Description

technical field [0001] The present invention relates to computer technology, in particular to a cloud computing implementation method and system. Background technique [0002] Traditional cloud computing implementation methods and systems implement cloud computing by virtualizing multiple physical servers and establishing virtual machines on the physical servers. In a traditional cloud computing system, each virtual machine implements different functions, and the operating systems and application programs configured on different virtual machines are usually different, and multiple virtual machines work together to realize corresponding functions. [0003] Because each virtual machine implements different functions, the operating systems and applications configured on different virtual machines are usually different. Therefore, in traditional cloud computing systems, when any virtual machine in the system fails, it will affect the operation of the entire system. , resulting ...

AI Technical Summary

Problems solved by technology

[0005] However, due to another design reason of the ext3 file system, fsck must compare the log file (journal) and the original data (meta-data) at the same time, so the file system integrity check time cannot be predicted, and the larger the file system, the larger the data The larger the amount, the more serious the situation, and the fsck check time may be hours, days, or even weeks

[0006] The Linux operating system in the traditional cloud computing system only realizes the conversion of the Linux operating system from the physical server to the virtualization platform, but does not change the inherent characteristics and defects of the ext3 file system itself. When any virtual machine in the system appears When the file system is read-only, that is, when a fault occurs on the virtual machine, it will affect the operation of the entire system, cause business suspension, and even cause the system to crash

Images