Windows and Linux Robustness Benchmarks With Respect to Application Erroneous Behavior

Karama Kanoun (1) , Yves Crouzet (1) , Ali Kalakech (1) , Ana-Elena Rugina (2)

1 LAAS-TSF - Équipe Tolérance aux fautes et Sûreté de Fonctionnement informatique
2 ESA - Agence Spatiale Européenne = European Space Agency

This chapter presents the specification of dependability benchmarks for general-purpose operating systems with respect to application erroneous behavior, and shows examples of benchmark results obtained for various versions of Windows and Linux operating systems. The benchmark measures are: operating system robustness (as regards possible erroneous inputs provided by the application software to the operating system (OS) via the application programming interface), the OS reaction, and restart times in the presence of faults. Two workloads are used for implementing the benchmark: PostMark, a file system performance benchmark for operating systems, and the Java Virtual Machine (JVM) middleware, a software layer on top of the OS allowing applications in Java language to be platform independent.

Mots clés: temps de réponse; operating system; Linux; Windows; Dependability benchmarking; étalonnage de la sûreté de fonctionnement; robustesse

Domaines: Informatique et langage [cs.CL]

Karama Kanoun, Yves Crouzet, Ali Kalakech, Ana-Elena Rugina. Windows and Linux Robustness Benchmarks With Respect to Application Erroneous Behavior. Karama Kanoun et Lisa Spainhower. Dependability Benchmarking for Computer Systems, IEEE Computer Society et WILEY, pp.227-254, 2008, 978-0-470-23055-8. ⟨hal-00761609⟩