The open and dynamic environment of Internet computing demands new software reliability technologies.How to efficiently and effectively build highly reliable Internet applications becomes a critical research problem.T...The open and dynamic environment of Internet computing demands new software reliability technologies.How to efficiently and effectively build highly reliable Internet applications becomes a critical research problem.This paper proposes a research framework for predicting reliability of individual software entities as well as the whole Internet application.Characteristics of the Internet environment are comprehensively analyzed and several reliability prediction approaches are proposed.A prototype is implemented and practical use of the proposed framework is also demonstrated.展开更多
Small and middle size high-performance computing clusters are very popular for various applications.How to utilize theaccumulated log data generated in the past to optimize job scheduling using machine learning techni...Small and middle size high-performance computing clusters are very popular for various applications.How to utilize theaccumulated log data generated in the past to optimize job scheduling using machine learning techniques is an interestingproblem.Most of the current work use the common machine learning algorithms,such as the multivariate linear regressionand polynomial model,to predict job runtime and optimize job scheduling.They either ignore the interference among jobfeatures or require a high time overhead for improving the prediction accuracy.In this paper,we propose to implement andimprove broad learning algorithm for predicting the execution times of new coming jobs more accurately and efficiently.Theexperimental results showed that the proposed method can obtain high prediction accuracy with a negligible time overhead.And the predicted job execution time can help improve the efficiency of job scheduling and HPC systems.展开更多
基金supported by the National Natural Science Foundation of China(Project No.61472338,61332010)Guangdong Natural Science Foundation(Project No. 2014A030313151)+1 种基金the Fundamental Research Funds for the Central Universitiesthe Research Grants Council of the Hong Kong Special Administrative Region,China (No.415113)
文摘The open and dynamic environment of Internet computing demands new software reliability technologies.How to efficiently and effectively build highly reliable Internet applications becomes a critical research problem.This paper proposes a research framework for predicting reliability of individual software entities as well as the whole Internet application.Characteristics of the Internet environment are comprehensively analyzed and several reliability prediction approaches are proposed.A prototype is implemented and practical use of the proposed framework is also demonstrated.
基金supported in part by Innovation Capability Support Program of Shaanxi(Program No.2022PT-10)the Fundamental Research Funds for the Central Universities.
文摘Small and middle size high-performance computing clusters are very popular for various applications.How to utilize theaccumulated log data generated in the past to optimize job scheduling using machine learning techniques is an interestingproblem.Most of the current work use the common machine learning algorithms,such as the multivariate linear regressionand polynomial model,to predict job runtime and optimize job scheduling.They either ignore the interference among jobfeatures or require a high time overhead for improving the prediction accuracy.In this paper,we propose to implement andimprove broad learning algorithm for predicting the execution times of new coming jobs more accurately and efficiently.Theexperimental results showed that the proposed method can obtain high prediction accuracy with a negligible time overhead.And the predicted job execution time can help improve the efficiency of job scheduling and HPC systems.
