A Hybrid Expansion Method for Virtual Machine Room

Abstract

The hybrid expansion method of a virtual machine room belongs to the technical field of virtual machine configuration, and includes the following steps: step S1, establishing a virtual machine room architecture: running an OpenStack system in a cloud virtual machine room; the OpenStack system is equipped with Nova components, Heat components, and Ceilometer components; Nova Elements, Heat elements, and Ceilometer elements implement the reactive automatic expansion mechanism of the OpenStack system; at the same time, the OpenStack system adds: administrator elements, forecaster elements and processor elements; step S2, execute the predictive expansion mechanism and OpenStack system's reactive auto-scaling mechanism and makes decisions. This solution implements both the predictive expansion mechanism and the reactive automatic expansion of the OpenStack system to properly allocate resources for the workload of the cloud computer room.

Description

technical field [0001] The invention belongs to the technical field of virtual machine configuration, and in particular relates to a hybrid expansion method of a virtual machine room. Background technique [0002] In the IT industry, the computer room generally refers to the place where telecommunications, Netcom, mobile, dual-line, electricity, and government or enterprises store servers and provide IT services for users and employees. Before the development of virtual technology, the configuration of computer room functions or computing power was mainly achieved through the increase or decrease of physical servers. [0003] With the development of virtualization technology, virtualization enables users to run multiple operating systems on one server at the same time. Generally speaking, a server system can be configured with multiple nodes, and each node runs multiple virtual machines (VirtualMachine, VM) at the same time, so as to provide each user with an independent op...

AI Technical Summary

Problems solved by technology

[0006] However, on the one hand, the behavior of the predictive scaling mechanism depends on the historical workload, and the historical workload needs to be collected after a period of time. If the number of collected samples is insufficient, prediction cannot be made, so the cloud virtual machine room will not be able to prepare resources in advance. Meet user needs

On the other hand, the predictive expansion mechanism cannot predict the future workload very accurately, so the pre-prepared resources may be too much or too little

Too few resources may fail to meet service level agreement commitments, and too many resources may increase service provider costs

[0007] The OpenStack system provides a reactive automatic expansion mechanism. Although this method does not consume additional computing resources of the system to regularly predict future workloads, its characteristic is to wait until the system load has reached a critical value before processing, so it is often processed. The time point is too late, which reduces the efficiency of processing user needs

Images