Commodity Technologies

 

The past few years have seen an unprecedented level of innovation and progress in commodity technologies. Three areas have been critical in this development: the Web, distributed objects, and databases. Each area has developed impressive and rapidly improving software artifacts. Examples at the lower level include Hypertext Markup Language (HTML), Hypertext Transfer Protocol (HTTP), MIME, Internet Inter-ORB Protocol (IIOP), Common Gateway Interface (CGI), Java, JavaScript, JavaBeans, Common Object Request Broker Architecture (CORBA), Common Object Model (COM), ActiveX, Virtual Reality Markup Language (VRML), object broker ORBs, and dynamic Java servers and clients. Examples at the higher level include collaboration, security, commerce, and multimedia technologies. Perhaps more important than these raw technologies is a set of open interfaces that enable large components to be quickly integrated into new applications.

We believe that computational grid environments can and should be constructed that incorporate these commodity capabilities in such a way as to achieve both high performance and high functionality.

One approach to this goal would be to use just a few of the emerging commodity technologies as ``point solutions.'' For example:

• VRML or Java3D could be used for scientific visualization.
• Web (including Java applets) frontends could provide convenient customizable interoperable user interfaces to high-performance facilities [22].
• The public key security and digital signature infrastructure being developed for electronic commerce could enable more powerful approaches to secure high-performance systems.
• Java could become a common scientific programming language.
• The universal adoption of Java DataBase Connectivity and the growing convenience of Web-linked databases could result in the growing importance of systems that link large-scale commercial databases with high-performance computing resources.
• The emerging ``object Web" (linking the Web, distributed objects, and databases) could encourage a growing use of modern object technology.
• Emerging collaboration and other distributed information systems could encourage new distributed work paradigms in place of traditional approaches to collaboration [30].

Our focus, however, is not on such point solutions but on exploiting the overall architecture of commodity systems for high-performance parallel or distributed computing. One might immediately raise the objection that over the past thirty years, many other major broad-based hardware and software developments have occurred--such as IBM business systems, UNIX, Macintosh and PC desktops, and video games--without any profound impact on high-performance computing software. We believe, however, that the emerging distributed commodity computing and information system (DcciS) is different: it gives us a worldwide, enterprise-wide distributing computing environment. Whereas previous software revolutions could help individual components of a high-performance computing system, DcciS can, in principle, be the backbone of a complete high-performance computing software system--whether it be for some global distributed application, an enterprise cluster, or a tightly coupled large-scale parallel computer.

To achieve this goal, we must add high performance to the emerging DcciS environment. This task may be extremely difficult; but, by using DcciS as a basis, we inherit a multi-billion dollar investment and what is, in many respects, the most powerful productive software environment ever built.