Measuring software aging effects through OS kernel instrumentation

Abstract

Software aging is a phenomenon defined as the continuing degradation of software systems during runtime, being particularly noticeable in long-running applications. Measuring aging effects is a very important step towards identifying the most significant causes of software aging, so rejuvenation mechanisms can be applied effectively. The literature has already discussed aging monitoring from a userspace perspective. In this paper we present an innovative study where we explore OS kernel instrumentation techniques to measure software aging effects. Kernel instrumentations are available in practically all modern operating systems and we show how this powerful mechanism can be used to monitor either application-specific or system-wide aging indicators. In addition to the theoretical study, we present numerical results obtained from controlled experiments.