Key Mechanism Of Large Scale Evaluation Cloud Platform: Task Management And Monitoring

In today's highly informational era,the Internet has a wide and in-depth impact on all levels of society.The development of the Internet to today has presented many technical fields,rapid knowledge update,large application scale,and complex application technology stack.It has higher and more comprehensive requirements for the theoretical knowledge and practical skills of IT practitioners.At the same time,the IT talent market also shows the characteristics of many types of jobs,rapid technological changes and high capacity requirements,which puts higher demands on the training of IT talents.The cultivation of IT talents or information technology talents has been related to the development of various industries and even the entire country.However,as one of the most important training methods for IT talents,software engineering professional teaching has various drawbacks in its practice.Due to the speed specified by the university's training system,the talent training system of colleges and universities tends to be slow to update,unable to keep up with the pace of Internet technology changes.The knowledge and skills they teach are often out of line with the requirements in production,and the employers need to continue to target Sexual training can only meet the job requirements.At the same time,due to the limited practical teaching resources of colleges and universities,more human resources are required in the process of assessment and task completion,resulting in lower teaching efficiency.The shortcomings in the practical teaching of software engineering affect the speed and quality of Internet talent training under the new requirements of the new era.Further,it also affects the country's talent supply to the Internet and information industry.At present,there are certain solutions to the problems in the practical teaching of IT personnel training.More typical are online programming platforms and online experimental platforms.However,existing online programming platforms mostly focus on the training of basic algorithms,lack of support for complex projects and advanced technologies,and users cannot obtain real project development experiences.The online experimental platform focuses on online software engineering experiments,lacking assessment of user learning results or time operations,users can not continue learning on the platform.In view of the problems in the current software engineering practice teaching and the current solutions,this paper proposes a cloud platform implementation method that can support large-scale complex projects and advanced technology online evaluation.Through the abstraction and refinement of the evaluation process,this topic realizes the automatic evaluation mechanism based on the virtualized container,uses the container as the evaluation running space,realizes the management of the container cluster with Kubernetes,and carries out the problems of security and load balancing in the automatic evaluation.Research has ensured that the platform has the ability to respond to large-scale concurrent evaluations.A large-scale evaluation task management method based on automatic evaluation mechanism is proposed.The evaluation process is optimized for user evaluation habits,and effective support for large-scale complex projects and advanced technology online evaluation is realized.In order to ensure the stable operation of the cloud platform and ensure that the system can be optimized and upgraded in time,this paper deeply analyzes the architecture and operation process of the entire evaluation platform,and through the analysis of the system operation data,through the correlation between abnormality and evaluation of service status.The valuable part is selected as the monitoring point from the many monitoring points,and based on the annotation mode,the evaluation request processing chain positioning mechanism based on the process division is realized,and the service request of the evaluation request in the distributed environment is realized.The situation is tracked,and mechanisms such as abnormal positioning,system early warning,system performance evaluation and diagnosis are realized based on the processing chain positioning mechanism.In the evaluation platform and monitoring prototype system implementation stage,through reasonable architecture design and business logic design,Kubernetes and Docker's underlying architecture was used to implement the evaluation cloud platform,and Kafka distributed message middleware and Storm stream were selected according to the requirements of the monitoring system.The processing framework implements the monitoring prototype system,and verifies the feasibility and value of the work through a series of experiments and analysis.