A Multi-Agent Framework for Testing Distributed Systems

Abstract

Software testing is a very expensive and time consuming process. It can account for up to 50% of the total cost of the software development. Distributed systems make software testing a daunting task. The research described in this paper investigates a novel multi-agent framework for testing 3-tier distributed systems. This paper describes the framework architecture as well as the communication mechanism among agents in the architecture. Web-based application is examined as a case study to validate the proposed framework. The framework is considered as a step forward to automate testing for distributed systems in order to enhance their reliability within an acceptable range of cost and time.

Unit Testing Non-functional Concerns of Component-based Distributed Systems

“…al [5] and Yamany et. al [22] describe techniques for unit testing multi-agent systems using so-called mock objects. Their goal for unit testing multi-agent systems is similar to UNITE, though they focus on functional concerns, whereas UNITE focuses on non-functional concerns of a distributed system during the early stages of development.…”

Agent Systems in Software Engineering

“…On the one hand, the application domain is related to web services, web applications and database applications. The only exception is the work of (Yamany et al, 2006) but even in this case only 3-tier applications are considered. These domains have a surely specific nature even though they provide a solid basis for introducing the existing attempts.…”

“…However, in both cases there is no actual system to verify the expected benefits of the two mechanisms. With respect to agent frameworks (Yamany et al, 2006;Bai et al, 2006;Kung, 2004;Miao et al, 2007), the common aspiration is to model software testing. The testing process is decomposed into phases during test planning while these plans can be executed asynchronously.…”

“…In (Yamany et al, 2006), a design for testing distributed systems is proposed. They use a three-tier distributed system structure consisting of a server, middleware and multiple clients.…”

Reliability Models Applied to Mobile Applications

“…Reliability is an important attribute of software quality in addition to other attributes such as performance, usability and fault prediction [5]. Software testing techniques can impact reliability measurement [6]. Software Reliability Growth Models (SRGMs) have been successfully applied to desktop (classic/standard) applications to assure high reliability, as was the case with Hardware Reliability Growth Models applicability to both desktops/laptops and mobile phone hardware [1].…”