Load regulation method and load regulation device in cloud computing environment

Abstract

The embodiment of the invention provides a load regulation method and a load regulation device in a cloud computing environment. The cloud computing environment comprises one or more storage physical machines, and computing examples operate on the storage physical machines. The method comprises the steps of computing the load state of the currently processed storage physical machine; when the condition that the storage physical machine is in a low-load state is judged, increasing the resource limit of the corresponding computing example; when the condition that the storage physical machine is in a high-load state is judged, decreasing the resource limit of the corresponding computing example; when the condition that the storage physical machine is in an overload state is judged, transferring the corresponding computing example to other storage physical machines, or, stopping operation of the corresponding computing example. The spare computing resources of the physical machines can be fully and reasonably utilized, the service quality is improved, isolation among tasks operating on the physical machines is improved, and excessive utilization of the resources of the physical machines by tasks beyond the storage service is limited, so that the influence of the tasks beyond the storage service on the storage service is reduced.

Description

technical field [0001] The invention relates to the technical field of cloud computing, in particular to a method for adjusting load in a cloud computing environment and a device for adjusting load in a cloud computing environment. Background technique [0002] With the development of Internet technology, cloud computing has become a key strategy for most enterprises and organizations. Clouds can be divided into three types: public cloud, private cloud and hybrid cloud. Cloud storage is a specific application of cloud computing. [0003] Storage services in private clouds are widely used. Internet applications can be roughly divided into the following categories: computing-intensive, network-intensive, and disk-intensive applications. In storage services, physical machines are usually used as the main storage unit, and each physical machine has multiple resources, such as CPU, memory, disk, network bandwidth, and network ports. [0004] However, storage services as a whol...

AI Technical Summary

Problems solved by technology

[0007] However, the storage service provider chooses a physical machine with relatively low performance as the back-end storage unit, and the low-performance physical machine will affect the performance of the storage service, and the service quality is low

[0008] On the other hand, storage service providers directly run some computing-intensive tasks on physical machines. Due to the poor isolation between tasks when running tasks directly, it will have a great impact on storage services, and this impact is unpredictable.

[0009] Finally, the actual business load on the physical machine is different in each time period, and the quota (usage and quota) of the physical machine resources cannot be adjusted in real time. If computing-intensive tasks occupy too many resources of the physical machine and cannot be adjusted, the storage The service fails to obtain the corresponding resources, which will affect the storage service. If the computing-intensive task occupies too few resources of the physical machine and cannot be adjusted, there will be unused computing resources in the physical machine, so that the computing resources will not be used reasonably.

Images