Research On Cloud Platform Fault Detection And Recovery Capability Evaluation System

Since the concept of cloud computing has been formally proposed,many academic scholars and well-known enterprises in the industry have invested a lot of resources in the research and practice of cloud computing.In recent years,many important research results have appeared in the field of cloud computing.With the deepening of research and application,new problems and demands have emerged.In the past ten years,there have been different cloud computing platform failure events in cloud service provision,even if it is like the Google and Amazon.The company cannot avoid the emergence of cloud platform failures.Cloud computing has changed the way traditional applications and services are deployed,so if the cloud platform fails,it will cause serious economic and reputational damage to many Internet companies.Because of such problems,the research on cloud computing platform fault detection and fault recovery technology has become a hot spot in recent years.There have been many research results of fault detection and recovery,but for so many cloud computing platform fault detection and recovery.There is almost no systematic research on the evaluation of the method.The main research content of this paper focuses on the cloud detection platform fault detection and recovery capability evaluation system research.This paper analyzes the characteristics of today's cloud computing platform,summarizes the current development status of fault injection,fault detection and fault recovery technology through a large number of reference data,and combines the common cloud computing platform fault types to design cloud computing platform fault detection and recovery.The capability assessment system focuses on the design of assessment indicators and assessment models.In order to verify the cloud platform fault detection and recovery capability evaluation system designed in this paper,based on the existing experimental environment,the openstack platform is selected to build the experimental test environment.The evaluation tool is designed in detail according to the evaluation requirements.The design process selects the appropriate index collection.And storage technology,fault injection technology and corresponding report generation technology.During the test,this paper selects common computing service faults,message queue service faults,API service faults,and CPU faults as typical faults to test the evaluation system.The research focus of this paper is on the design of indicators and the design of evaluation models.In terms of fault detection index design,this paper uses fault detection time as the main indicator to measure fault detection capability.In terms of fault recovery index design,this paper divides the fault recovery methods into high-availability strategy and virtual machine migration strategy.For high-availability strategy,this paper selects data recovery time and data recovery degree as the main evaluation indicators.For virtual machine migration strategy,this paper selects Total migration time and downtime were used as the main indicators of assessment.In terms of the evaluation model,this paper mainly adopts the normalization of the indicator data,and displays the normalized data in the form of a radar chart.In this paper,the metric data obtained by fault detection and fault recovery experiments are aggregated into the capability evaluation values of this paper by data aggregation to evaluate the fault detection and recovery capabilities of the cloud platform.