Main effects screening: a distributed continuous quality assurance process for monitoring performance degradation in evolving software systems

Developers of highly configurable performance-intensive software systems often use a type of in-house performance-oriented "regression testing" to ensure that their modifications have not adversely affected their software's performance across its large configuration space. Unfortunately, time and resource constraints often limit developers to in-house testing of a small number of configurations and unreliable extrapolation from these results to the entire configuration space, which allows many performance bottlenecks and sources of QoS degradation to escape detection until systems are fielded. To improve performance assessment of evolving systems across large configuration spaces, we have developed a distributed continuous quality assurance (DCQA) process called main effects screening that uses in-the-field resources to execute formally designed experiments to help reduce the configuration space, thereby allowing developers to perform more targeted in-house QA. We have evaluated this process via several feasibility studies on several large, widely-used performance-intensive software systems. Our results indicate that main effects screening can detect key sources of performance degradation in large-scale systems with significantly less effort than conventional techniques.

1. Introduction

The quality of performance-intensive software systems, such as high-performance scientific computing systems and distributed real-time and embedded (DRE) systems, depend heavily on their infrastructure platforms, such as the hardware, operating system, middleware, and language processing tools. Developers of such systems often need to tune the infrastructure and their software applications to accommodate the (often changing) platform environments and performance requirements. This tuning is commonly done by (re)adjusting a large set (10's-100's) of compile- and runtime configuration options that record and control variable software parameters, such as different operating systems, resource management strategies, middleware and application feature sets; compiler flags; and/or run-time optimization settings. For example, SQL Server 7.0 has 47 configuration options, Oracle 9 has 211 initialization parameters, and Apache HTTP Server Version 1.3 has 85 core configuration options.