An Effective Algorithm and Modeling for Information Resources Scheduling in Cloud Computing

With the development of the Internet and Web technology and frequent cross-cultural communication, it is necessary for people to be able to access and manage information in many different languages. The scheduling of multilingual information resources become complex, as multilingual information resources stored in different cloud datacenters are heterogeneous and their distributions are uneven. In this paper, based on Min-Min algorithm, Least Languages First Min-Min (LLFMM), a scheduling algorithm that is suitable to multilingual information resources and can improve information resources scheduling performance in cloud computing is proposed. A task with least number of sites that can fulfill this task is firstly selected in LLFMM, rather than the task with the least execution time among all the tasks in Min-Min. Petri net is a powerful graphical and mathematics tool for describing the concurrent, asynchronous and dynamic events. This paper models and analyzes the multilingual information resources scheduling using colored dynamic timed Petri nets (CDTdPN). In CDTdPN model, the color represents different language information resource, and the time of a transition representing scheduling time is a function of tokens in input places of the transition. The concurrent scheduling marking graph (CSMG) of CDTdPN is defined and the algorithm for constructing the CSMG of CDTdPN is given in this paper. The CSMG can represent the concurrent relation of transitions in Petri nets and reduce the size of the graph to be constructed in some extent. Finally, the example and simulation show that our scheduling algorithm is effective in reducing makespan and our modeling and analyzing techniques are convenient for analyzing performance of scheduling system.