Cluster monitoring system and monitoring method for multi-public cloud computing platforms

Abstract

The present invention provides a cluster monitoring system for multi-cloud computing platforms, including a data acquisition subsystem responsible for collecting basic resource data, task operation status and consumption, and overall resource usage status monitoring data of each cloud computing platform according to predetermined indicators, and Provide an interface for the scheduling system to call to obtain real-time monitoring data to guide scheduling; the data processing subsystem is responsible for the cluster monitoring data uploaded by each data subsystem through RPC requests, performs a series of processing dumps to the back end for storage, and provides interfaces for data Display and billing system; the alarm subsystem is responsible for processing and analyzing monitoring data according to the alarm strategy, confirming the alarm level and sending alarm information. The present invention can better observe the overall task operation situation, improve the resource utilization rate, facilitate bill audit and reconciliation with the persistence of computing resources, and can automatically solve some abnormal problems and reduce manual participation.

Description

technical field [0001] The invention belongs to the technical field of cross-multiple public cloud computing scheduling, and specifically relates to a multi-public cloud computing platform-oriented cluster monitoring system and a monitoring method thereof, which can be used in multi-cloud computing platform clusters as multi-computing cluster resource states, computing Data support system for task status monitoring alarm system and other related systems. Background technique [0002] Cloud computing is a pay-by-use model that provides available, convenient, and on-demand network access and access to a configurable pool of computing resources (resources include networks, servers, storage, application software, and services) , these resources can be provisioned quickly with little administrative effort or interaction with the service provider. [0003] With the development of technologies such as the Internet and cloud computing, more and more public cloud computing resources...

AI Technical Summary

Problems solved by technology

At present, none of the cloud providers can complete the support for these monitoring indicators

[0007] 2. Based on cost considerations, cloud providers cannot customize the collection of monitoring indicators and cannot conduct more detailed analysis of monitoring data, and thus cannot feedback the scheduling system to adjust strategies in real time

In 1, it is explained that the cloud provider cannot provide some monitoring indicators required by the cloud computing platform, and the cloud provider cannot allow users to collect and display these indicators in a certain way

Existing monitoring data is stored on the cloud and there is no way to obtain historical data for a certain period of time for more detailed analysis, and then it is impossible to check resource usage, resulting in the inability to check resource usage bills

If the data that the scheduling system relies on for policy adjustment cannot be obtained, real-time scheduling policies cannot be discussed, and a single scheduling policy may cause unnecessary waste of resources

[0008] 3. The built-in resource monitoring view is scattered and cannot provide a user-customizable unified view, and the monitoring methods of each cloud provider are different, so it is difficult to have a unified way to process the resource monitoring data of each cloud

Cloud provision not only fails to provide collection of certain indicators, but also makes it difficult to integrate basic resources, which also fails to provide an intuitive view of resource status

[0009] 4. The alarm system can only be set separately for one or a certain type of resource and cannot classify the alarms

Cloud providers can provide threshold alarms for basic resources, but they cannot classify these alarms. When there are too many alarm information, it is easy to cause key alarm information to be ignored and cannot be processed in time.

Moreover, the existing alarm strategy cannot meet the alarm requirements of the cloud computing platform at all, so it is impossible to fully grasp the runtime status of the computing platform

Images