Carnegie Mellon University

File(s) stored somewhere else

Please note: Linked content is NOT stored on Carnegie Mellon University and we can't guarantee its availability, quality, security or accept any liability.

Enabling Predictable Assembly

journal contribution
posted on 2003-03-01, 00:00 authored by Scott A. Hissam, Gabriel A. Moreno, Judith Stafford, Kurt C. Wallnau
Demands for increased functionality, better quality, and faster time-to-market in software products continue to increase. Component-based development is the software industry’s response to these demands. The industry has developed technologies such as EJB and CORBA to assemble components that are created in isolation. Component technologies available today allow designers to plug components together, but do little to allow the developer to reason about how well they will play together. Predictable assembly focuses on issues related to assembling component-based systems that predictably meet their quality attribute requirements. This paper introduces prediction-enabled component technology (PECT) as a means of packaging predictable assembly as a deployable product. A PECT is the integration of a component technology with one or more analysis technologies. Analysis technologies support prediction of assembly properties and also identify required component properties and their certifiable descriptions. This report describes the major structures of a PECT. It then discusses the means of validating the predictive powers of a PECT, which provides measurably bounded trust in design-time predictions. Last, it demonstrates the above concepts in an illustrative model problem: predicting average end-to-end latency of a ‘soft’ real time application built from off-the-shelf software components.




Usage metrics


    Ref. manager