Study on resource service match and search in manufacturing grid system

Abstract

Resource service match and search (RSMS) is the core to realize manufacturing grid (MGrid) resource scheduling. In order to realize effectively RSMS between resource demanders and providers, a RSMS framework is proposed and the key technologies to realize it are studied. The describing information of resource services are classified into four categories: (a) word concept information, (b) sentence information, (c) number information, including number interval and fuzzy number, and (d) entity class (or data structure) information. The similarity matching algorithms of each kind of describing information are investigated, respectively, including word matching algorithms, sentence matching algorithms, number matching algorithms, and entity class matching algorithms. Based on the proposed matching algorithms, the match and search processes of MGrid resource services are divided into four phases: first, matching the basic information of resource services, such as service name and service description, namely, basic matching; second, matching the inputs and outputs information of resource services, namely, I/O matching; third, matching the quality of service (QoS) information of resource services, namely QoS matching; last, combining the above three matching results and generating an integrated matching result, namely, integrated matching. The matching functions and algorithms of each phase are described in detail. A case study illustrates the application of proposed methods, and the accuracy and efficiency of the proposed method are measured.