In this chapter, we consider the role of commodity, off-the-self software technologies and components in the construction of computational grids. We take the position that computational grids can and should build on emerging commodity network computing technologies, such as CORBA, Microsoft's COM, JavaBeans, and less sophisticated Web and networked approaches. These technologies are being used to construct three-tier architectures, in which middle-tier application servers mediate between sophisticated backend services and potentially simple frontends. The decomposition of application functionality into separate presentation, application, and backend service tiers results in a distributed computing architecture that, we argue, can be extended transparently to incorporate grid-enabled second- and/or third-tier services. Consequently, the three-tier architecture being deployed for commodity network applications is well suited to serve as an architecture for computational grids, combining high performance with the rich functionality of commodity systems. The distinct interface, server, and specialized service implementation layers enable technological advances to be incorporated in an incremental fashion.
This commodity approach to grid architecture should be contrasted with
more specialized grid architectures such as Globus
(Chapter 
) and Legion
(Chapter ). Clearly, these latter technologies
can be used to implement lower-tier services. This would seem to be
consistent with the Globus design philosophy, given that it defines a
service-oriented approach to metacomputing. The integration of Legion
into the commodity architecture is more problematic, in that Legion
defines a model for the programmer that the three-tier model will
not expose. That said, Legion could be encapsulated into backend
services as well.
The rest of this chapter proceeds as follows. We first define what we mean by commodity technologies and explain the different ways that they can be used in high-performance computing. Then, we discuss an emerging distributed commodity computing and information system in terms of a conventional three-tier commercial computing model. We describe how this model can be used as a CORBA facility, and we give various examples of how commodity technologies can be used effectively for computational grids. Finally, we discuss how commodity technologies such as Java can be used to build parallel programming environments that combine high functionality and high performance.