Lodestone: A Streaming Approach to Behavior Modeling and Load-Testing

EasyChair Preprint 2743

Abstract

It is evident that our technologically-dependent society rightly expects systems engineers to produce systems having increasing levels of performance, efficiency, and reliability. As such, we must convolve academic and industrial approaches to provide theory, systems, and working technologies which can catalyze and propel engineers, developers, and technical professionals of various disciplines toward the ultimate goal of consistent delivery of quality systems. There exist tools and processes for improving the quality-oriented posture of the systems engineering industry; in practice, the most perpetual form of software testing continues to be rote repetition of test cases through either manual testing or scripted automation of those same manual tests.

We describe Lodestone: a real-time approach for generating workload in software systems. This real-time approach to LT uses streaming log data to generate and dynamically update user behavior models, cluster them into similar behavior profiles, and instantiate distributed workload of software systems. We show that Lodestone outperforms Markov4JMeter through a qualitative comparison of key feature parameters as well through experimentation based on shared data and models.

Keyphrases: Automatic Test Pattern Generation, Software Performance, automatic testing, performance analysis, software quality, software testing, system performance, system testing

@booklet{EasyChair:2743,
  author    = {Chester Parrott and Doris Carver},
  title     = {Lodestone: A Streaming Approach to Behavior Modeling and Load-Testing},
  howpublished = {EasyChair Preprint 2743},
  year      = {EasyChair, 2020}}