Deriving Scalability Tactics for Development of Data-Intensive Systems

Abstract

Enforcing scalability quality attribute to an existing developed data-intensive software system is a difficult task. This paper proposes a methodology that reverses the traditional procedure by reframing scalability tactics. The scalability tactics are derived from existing scalable data-intensive system development practices and are presented in the format of a utility tree. These tactics can be analyzed against architectural patterns which are design decessions for software development. This will maximize the hold of scalability attributes in end software product before its actual development. The tactics derived in this paper will help an architect to have the correct design decision for data-intensive system development. Tactics alone may not solve the problem but when contrasted with architectural patterns, they will help in determining if the system is scalable.