虚拟计算环境下基于聚类的资源匹配优化模型

发布时间：2018-05-18 00:08

本文选题：任务聚类 + 二分K均值　；参考：《北京邮电大学学报》2017年S1期

【摘要】：虚拟计算环境中任务具有数量庞大、需求模糊、种类多样等特征,使得资源匹配面临巨大挑战.依据虚拟计算实验床平台公布数据,提出了一种融合虚拟资源与任务聚类的资源匹配优化模型.该模型通过分析任务需求、消耗等特征,基于改进二分K均值进行任务聚类,并结合虚拟资源类型生成优化的资源匹配列表.经实验分析验证,该模型有效缩小资源匹配范围,提高任务运行成功率,为精准匹配提供基础.
[Abstract]:In virtual computing environment, resource matching is faced with enormous challenges due to the large number of tasks, fuzzy requirements, diverse types and so on. According to the published data of virtual computing experimental bed, a resource matching optimization model combining virtual resource and task clustering is proposed. By analyzing the characteristics of task requirements and consumption, the model clusters tasks based on the improved binary K-means, and generates an optimized resource matching list based on virtual resource types. The experimental results show that the model can effectively reduce the range of resource matching, improve the success rate of task operation, and provide the basis for accurate matching.
【作者单位】：北京邮电大学信息安全中心;国家计算机网络应急技术处理协调中心;
【基金】：国家242信息安全计划项目(2015A136)
【分类号】：TP311.13

【相似文献】