Grid Services for Distributed System
Ian T Foster, Steven Tuecke · 2002
abstractions and concepts that let applicationsaccess and share resources and services across dis-tributed, wide area networks, while providing com-mon security semantics, distributed resourcemanagement performance, coordinated fail-over,problem determination services, or other QoS met-rics that are of importance in a particular context.For some time, such problems have been of cen-tral concern to developers of distributed systems forlarge-scale scientific research. Work within thiscommunity has led to the development of Grid tech-nologies , 1 which have been widely adopted in sci-entific and technical computing. 2 Grid technologiesand infrastructures support the sharing and coor-dinated use of diverse resources in dynamic, dis-tributed virtual organizations 3 —that is, thecreation, from geographically distributed compo-nents operated by distinct organizations with dif-fering policies, of virtual computing systems thatare sufficiently integrated to deliver the desired QoS.In particular, the open source Globus Toolkitdescribed in the “OGSA and the Globus ToolKit”sidebar has emerged as a de facto standard for con-struction of Grid systems. Projects building on theGlobus Toolkit range from scientific collaborationsconcerned with remote access to specialized exper-imental facilities—for example, the Network forEarthquake Engineering Simulation, NEESgrid;(http://www.neesgrid.org)—to “data grids” for the