Operating System Robustness Forecast and Selection

Ju, Xiaoen; Zou, Haiyuan

doi:10.1109/issre.2008.10

Cited by 7 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SWIFI techniques test the tolerance of the system through the simulation software or hardware errors. The Ballista Project injects faults with corrupting parameters of POSIX API calls [8] [13] which have proved to be an effective way to assess the robustness of general operating systems. Xception [9] injects faults into code with extra exception code to simulate hardware faults.…”

Section: Related Workmentioning

confidence: 99%

A Workload Model Based Approach to Evaluate the Robustness of Real-time Operating System

Zhou

Cai

et al. 2013

2013 IEEE 10th International Conference on High Performance Computing and Communications &Amp; 2013 IEEE International Conferen

View full text Add to dashboard Cite

An important step in the development of real-time operating system (RTOS) is the validation of its fault tolerance properties. Many efforts have been made to improve the robustness of RTOS. However, most of the robustness tests are not complete as the influence of workload on these tests is ignored. In this paper, faults are injected through a set of public application program interfaces (APIs) which are based on the basic components of RTOS. A comprehensive evaluation system which combines the general RTOS API robustness with the complex workloads is proposed and three popular RTOSs (Ucos2.62, Vxwork5.4 and Rtems4.10) have been evaluated with two workloads by this evaluation system. We find Rtems shows the best performance while Vxworks performs the worst.

show abstract

Section: Related Workmentioning

confidence: 99%

A Workload Model Based Approach to Evaluate the Robustness of Real-time Operating System

Zhou

Cai

et al. 2013

2013 IEEE 10th International Conference on High Performance Computing and Communications &Amp; 2013 IEEE International Conferen

View full text Add to dashboard Cite

show abstract

“…Software reliability has always been a concern for software developers, practitioners, and users [1,2,3]. Many factors affect software reliability: the skill level of the developers, the mature level of the development process, and the platform used for software development/execution such as programming language, compiler, operating system, and instruction set architecture [4].…”

Section: Introductionmentioning

confidence: 99%

Cross Comparison on C Compilers' Reliability Impact on UNIX Based Fedora OS

Karna

Zou

2010

2010 10th IEEE International Conference on Computer and Information Technology

Self Cite

View full text Add to dashboard Cite

Software reliability has always been a concern for both software producers and consumers. Because of their role in binary code generation, compilers have significant impact on software performance. Although many works have been conducted on compilers, little is done on their impact on software reliability. Thus, people often face a void of useful knowledge when deciding which compiler to use in developing software to meet a specific reliability target.This paper tries to provide some guidance in users' compiler selection by performing cross-comparison study on popular C compilers: GCC, Intel, Sun, PGI, LLVM using the UNIX based Fedora operating system. We have used SPEC CPU2006 benchmark suite as our test set. Our results show that on average, GCC C compiler is the best while the PGI C compiler is superior to Intel on this operating system platform. And this result is consistent with user experience from our survey feedback.

show abstract