Design and Implementation of Artificial Intelligence Experimental Platform Based on Serverless Architecture
Aiming at the practical problems of artificial intelligence experimental teaching in complex experimental environ-ment,large amount of experimental data and high difficulty of algorithm implementation,an artificial intelligence experimental plat-form based on Serverless architecture is designed and implemented.Serverless architecture is designed and built by integrating con-tainer construction,workload management and event triggering.The platform uses Docker image as the image of the experimental container,in which the scikit-learn algorithm library and a large number of public data sets are integrated to complete the migration of massive artificial intelligence basic algorithms.The experimental teaching administrators add new experiments in the form of sub-mitting experimental code packages.Students can use a large number of common algorithms in the experimental container to com-plete experiments,which improves the extensibility and usability of the platform.At the same time,the platform based on Serverless architecture to achieve the experimental container dynamic expansion capacity,increase the platform's multi-user concurrency abili-ty.
国家科技期刊平台
NSTL
万方数据
