Data grid and gridflow management systems

Abstract

The 'Grid' is an emerging infrastructure providing coordinated and consistent services to access distributed, heterogeneous computational and information storage resources amongst autonomous organizations. Data grids are being built across the world as the next generation inter-organizational data management systems for coordinated sharing of data as well as storage resources. A data grid is a location independent logical namespace consisting persistent global identifiers for digital entities, storage resources and users in an autonomous inter-organizational collaboration. Data Grid Management Systems (DGMS) provide services on the data grid?s logical name space for management and organization of digital entities and resources. These data-intensive environments involve long run processes, which could be aggregated as Grid Workflows (Gridflows). Gridflows are executed as peer-to-peer workflows in the data grid infrastructure by the Gridflow Management Systems. The tutorial will introduce the grid, provide use-cases in large projects, examine requirements, introduce some possible solutions for managing data in the grid and using a services-oriented infrastructure. It will take real-world examples from current systems, in particular the Storage Resource Broker and the SDSC Matrix project.