异构混合云服务下的多任务算力度量方法

冯汉枣; 黎元宝; 刘运奇

异构混合云服务下的多任务算力度量方法

引用本文：冯汉枣，黎元宝，刘运奇.异构混合云服务下的多任务算力度量方法[J].计算技术与自动化,2023,(4):154-158

摘要点击次数: 145

全文下载次数: 0

作者	单位
冯汉枣，黎元宝，刘运奇	（联通（广东）产业互联网有限公司，广东广州 510320）

中文摘要:异构网络中混合云服务下的任务数据量复杂程度较高、数据维度高，影响多任务算力度量的效率及完成率，提出了异构混合云服务下的多任务算力度量方法。分析异构混合云服务系统，获取多任务算力数据，依据获取结果从中提取多任务算力特征集。利用构建的度量模型对多任务算力特征进行标识及量化，通过对模型及特征的实例相似性计算，建立多任务算力度量矩阵，根据该矩阵实现多任务算力度量。实验结果表明：通过对该方法实施度量性能及度量效果对比测试，验证了该方法对任务算力度量的耗时短，多任务算力度量完成率较高。

中文关键词:异构混合服务云多任务算力度量方法特征集标识及量化

Multi-task Computing Power Measurement Under Heterogeneous Hybrid Cloud Services

Abstract:The complexity and dimension of task data under hybrid cloud service in heterogeneous networks are high, which affects the efficiency and completion rate of multi task computing power measurement. A multi task computing power measurement method under heterogeneous hybrid cloud service is proposed. Analyzing the heterogeneous hybrid cloud service system, obtaining the multitask computing power data, and extracting the multitask computing power feature set according to the obtained results. Using the constructed measurement model to identify and quantify the characteristics of multitask computing power, and establishing a multitask computing power measurement matrix by calculating the similarity of the model and the examples of the features, and then implementing the multitask computing power measurement according to the matrix. The experimental results show that the method takes less time to measure the task computing power and has a higher completion rate of multitask computing power measurement through the comparison test of the measurement performance and measurement effect of the method.

keywords:heterogeneous hybrid service cloud multi-task computing power measurement method feature set identification and quantification

查看全文 查看/发表评论 下载pdf阅读器