Title :
Operating System Robustness Forecast and Selection
Author :
Ju, Xiaoen ; Zou, Hengming
Author_Institution :
Dept. of Comput. Sci., Shanghai Jiao Tong Univ., Shanghai
Abstract :
While commercial off-the-shelf (COTS) operating systems (OS) have long been widely used, the issue regarding their robustness is far from being solved. Although many efforts have been made in this research domain, people still find it difficult to make choices among various OSs for robustness concerns. This paper proposes a reference model for OS robustness forecast and selection that aims to forecast the robustness of specific OSs under given operational profiles. At the same time, the model can select appropriate OSs as development/operating platforms that meet the particular robustness requirements of the target workloads. Our model combines OSspsila overall robustness with operational profiles and uses extensive tests on OS APIs to make our calculation. We have measured 255 APIs and C-library functions on windows XP and Vista, and 197 C-library functions on Linux 2.6.22 (Ubuntu 7.10). Our results show that on average Windows XP and Vista are more robust than Linux, but their performances are comparable under compute-intensive workloads. A demonstration of how these results are used in the proposed reference model for OSs robustness forecast and selection is given at the end.
Keywords :
application program interfaces; operating systems (computers); software libraries; software packages; C-library function; COTS; OS API; Vista; commercial Off-the-Shelf; operating system robustness forecast; operational profile measurement; selection model; windows XP; Application software; Computer science; Frequency; Linux; Operating systems; Predictive models; Reliability engineering; Robustness; Software reliability; Systems engineering and theory; OS robustness; operational profile; robustness forecast; robustness selection;
Conference_Titel :
Software Reliability Engineering, 2008. ISSRE 2008. 19th International Symposium on
Conference_Location :
Seattle, WA
Print_ISBN :
978-0-7695-3405-3
Electronic_ISBN :
1071-9458
DOI :
10.1109/ISSRE.2008.10