云计算环境下基于用户行为的资源调度机制的研究

项目名称： 云计算环境下基于用户行为的资源调度机制的研究

项目编号： No.61300176

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 自动化技术、计算机技术

项目作者： 丁丁

作者单位： 北京交通大学

项目金额： 23万元

中文摘要： 作为互联网资源利用的新模式，云计算的商业性使其调度机制更加关注用户的需求。针对云计算资源调度这种面向服务、面向用户的特点，本项目从用户行为的角度着手，利用用户行为所体现出的相关性和稳定性，深层挖掘用户的真实意图，试图找到提高资源调度效率、满足用户需求的调度机制，从而为具有不同需求的用户提供有保障的服务。主要研究内容包括两方面，一是通过借鉴推荐系统中的协同过滤技术，分析、提取用户行为来预测用户的真实目的，研究按需聚合的资源聚类模型和聚类方法，以减少资源搜索空间，提高资源调度的效率和准确性；另一个是利用用户交互行为，将用户行为信息融入到资源调度过程中，并结合相关反馈机制，研究面向用户需求的资源调度机制，实现用户需求的主动发现和自动调节。本项目的研究为云计算环境下资源调度问题的求解提供新的思路和方法。

中文关键词： 云计算；资源调度；用户行为；资源聚类；行为反馈

英文摘要： Scheduling mechanism in cloud computing environment concerns more about the needs of users due to the commercial purpose of cloud computing. In response to the service-oriented and user-oriented features, this project deep dives the problem of cloud resource scheduling based on user behavior and aims to find out a scheduling mechanism which can not only increase the efficiency of resource scheduling but also provide guaranteed service for users with devise requirements. The detail work includes two parts. First, by introducing the collaborative filtering technology learned from recommendation system, an associated model between user behavior and resource is built to predict the users' real purpose and an on-demand resource clustering method is proposed to reduce the range of search, improving the efficiency and accuracy of resource scheduling in cloud computing. Secondly, by combining with the relevance feedback mechanism, user behavior information collected from the interaction is integrated into the resource scheduling and a user requirement oriented resource scheduling mechanism is developed with the capability of active discovering and automatically adjusting of the needs of users. The research work above can provide a new idea and approach for the resource scheduling in cloud computing environment.

英文关键词： cloud computing；resource scheduling；user behavior；resource clustering；behavior feedback