Elastic resource forecasting and building method for cloud computing

Abstract

The invention discloses an elastic resource forecasting and building method for cloud computing. The method comprises the following steps: (1) a forecasting unit and a resource building unit are initialized; (2) the forecasting unit calculates the expectation, namely a formula shown in a drawing, of a first time point t1; (3) the resource building unit calculates the resource number M1 of the t1, wherein M1 is equal to a formula shown in a drawing; (4) the resource building unit selects a computing resource combination, and a cloud-computing virtual machine computing unit is started up; (5) whether a next time point t(n+1) reaches or not is judged; (6) if the t(n+1) reaches, the forecasting unit calculates the expectation, namely a formula shown in a drawing, of the t(n+1); (7) the resource building unit calculates the resource number M(n+1) of the t(n+1), wherein M(n+1) is equal to a formula shown in a drawing; and (8) the resource building unit compares the M1 with the M(n+1), the computing resource combination is increased when the M(n+1) is greater than the M1, when the M(n+1) is smaller than the M1, the computing resource combination is closed, and the operation is returned to the step (3). The method has the advantages that the effectiveness ratio of the cloud computing is increased, the characteristic that the cloud computing services on demand and is built elastically is effectively utilized, the effects of energy conservation and emission reduction are improved, and the like.

Description

technical field [0001] The present invention relates to a computing resource prediction and elastic construction method technology based on a cloud computing system platform, in particular to a cloud computing elastic resource prediction and construction method, which is based on a Poisson random process model and can meet the needs of Internet services Predicting and elastically scaling the computing resource scale of cloud computing virtual machines is a stochastic process mathematical model and a method for adjusting and constructing cloud computing virtual machine computing units. Background technique [0002] Network services have become an important foundation for the development of current social informatization. Cloud computing is one of the most important IT technologies after personal computers and the Internet. In the field of cloud computing infrastructure service technology, facing the construction of clustered resources, the on-demand service and elastic constr...

AI Technical Summary

Problems solved by technology

[0003] However, the existing cloud computing platform adopts a fixed resource pool method when constructing resources. The resource pool is constructed according to the maximum computing demand of the service provided, and the fixed resource is constructed in advance according to the maximum service demand to meet the virtual machine scale required by the network service. However, the network service demand of most services will not remain at the peak state for a long time, so adopting this method will cause the cloud computing resource utility ratio to be unable to be optimized

Images