Analysis of software halts in the tandem GUARDIAN operating system

Lee, I.; Iyer, K. B. Priya

doi:10.1109/issre.1992.285887

Cited by 14 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tang et al [37,36] and Buckley et al [5,6] have investigated error/failure logs collected from various VAXcluster systems of different sizes. Lee et al [19] and Lin et al [20] analyzed the error trends for Tandem systems and DCE environments. Xu et al [40] performed a study of error logs collected from a heterogeneous distributed system consisting of 503 PC servers.…”

Section: Failure Injectionmentioning

confidence: 99%

“…This includes the fitting of failure data to Weibull, lognormal and other specific distributions, each with different parameter settings, under the assumption of independent and identically distributed failures [20,19,13]. Other studies have demonstrated that the sequence of failures on some computer systems are correlated in various ways and that the failures tend to occur in bursts [37,36,40,28].…”

Section: Failure Injectionmentioning

confidence: 99%

See 1 more Smart Citation

Performance Implications of Failures in Large-Scale Cluster Scheduling

Zhang

Squillante

Sivasubramaniam

et al. 2005

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

Abstract. As we continue to evolve into large-scale parallel systems, many of them employing hundreds of computing engines to take on mission-critical roles, it is crucial to design those systems anticipating and accommodating the occurrence of failures. Failures become a commonplace feature of such large-scale systems, and one cannot continue to treat them as an exception. Despite the current and increasing importance of failures in these systems, our understanding of the performance impact of these critical issues on parallel computing environments is extremely limited. In this paper we develop a general failure modeling framework based on recent results from large-scale clusters and then we exploit this framework to conduct a detailed performance analysis of the impact of failures on system performance for a wide range of scheduling policies. Our results demonstrate that such failures can have a significant impact on the mean job response time and mean job slowdown under existing scheduling policies that ignore failures. We therefore investigate different scheduling mechanisms and policies to address these performance issues. Our results show that periodic checkpointing of jobs seems to do little to ease this problem. On the other hand, we demonstrate that information about the spatial and temporal correlation of failure occurrences can be very useful in designing a scheduling (job allocation) strategy to enhance system performance, with the former providing the greatest benefits.

show abstract

Section: Failure Injectionmentioning

confidence: 99%

Section: Failure Injectionmentioning

confidence: 99%

Performance Implications of Failures in Large-Scale Cluster Scheduling

Zhang

Squillante

Sivasubramaniam

et al. 2005

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

show abstract

“…The focus of such studies is on categorizing the nature of failures in the systems (e.g., software failures, hardware failures), identifying availability bottlenecks, and obtaining models to estimate the availability of the systems being analyzed. Lee [15], [16] analyzed failures in Tandem's GUARDIAN operating system. Tang [25] analyzed error logs pertaining to a multicomputer environment based on VAX/VMS cluster.…”

Section: Error and Failure Analysis Of A Lan Of Windows Nt-based Serversmentioning

confidence: 99%

“…errors that manifest themselves in multiple ways) was constructed to analyze system failure behavior. Measurement-based software reliability models were also presented in [15], [16] (for the GUARDIAN system) and [25], [26] (for the VAX cluster).…”

Section: Error and Failure Analysis Of A Lan Of Windows Nt-based Serversmentioning

confidence: 99%

Measurement-Based Analysis of System Dependability Using Fault Injection and Field Failure Data

Iyer

Kalbarczyk

2002

Performance Evaluation of Complex Systems: Techniques and Tools

View full text Add to dashboard Cite

Abstract. The discussion in this paper focuses on the issues involved in analyzing the availability of networked systems using fault injection and the failure data collected by the logging mechanisms built into the system. In particular we address: (1) analysis in the prototype phase using physical fault injection to an actual system. We use example of fault injection-based evaluation of a software-implemented fault tolerance (SIFT) environment (built around a set of self-checking processes called ARMORS) that provides error detection and recovery services to spaceborne scientific applications and (2) measurement-based analysis of systems in the field. We use example of LAN of Windows NT based computers to present methods for collecting and analyzing failure data to characterize network system dependability. Both, fault injection and failure data analysis enable us to study naturally occurring errors and to provide feedback to system designers on potential availability bottlenecks. For example, the study of failures in a network of Windows NT machines reveals that most of the problems that lead to reboots are software related and that though the average availability evaluates to over 99%, a typical machine, on average, provides acceptable service only about 92% of the time.

show abstract

“…Based on past analyses of failure data from field software systems [ 18. 29,38,39,42,43], the following operations are often rarely executed and are likely to cause problems in the operational phase: 9 redundancy management exception handling * initialization and calibration processing of data in boundary conditions or out of expected ranges…”

Section: Identification Of Critical System Modes and Their Occurrencementioning

confidence: 99%

An approach to measuring and assessing dependability for critical software systems

Tang

Hecht²

Proceedings the Eighth International Symposium on Software Reliability Engineering

View full text Add to dashboard Cite

Traditional software testing methods combined krsith probabilistic models cannot nieasure arid assess dependability for software that requires v e n high reliabilio (failure rate < 10-'/hour) and availabilit?. (>0. 999999). This paper proposes a novel approach, drawing on findings and methods that have been described individriull~ hut have never been combined, applied in the late testing phase or early operational phase, to quantifi depetitlubility for U categon of critical sofhvare with such high requirenietits. The concepts that are integrated ure: operational profle, rare conditions, importance sampling, stress testing, and measurement-based dependabilir). evaluation. In the approach, importance sanipling is applied on the operational profile to guide the testing of critical operations of the software, thereby accelerating the occurrence of rare conditions which have been shown to be a leading cause of failure in critical systems. The failure rates measured in the testing are then transformed to those that would occur in the normal operation by the likelihood ratio function of the importance sanipling theory, and finally depetidabilih for the tested software system is evaluated by using measurement-based dependabilio modeling techniques. When the acceleration factor is large (over loo), which is r)pical for a category of software of interest, it is possible to quantib a very high reliability or availability in a reasonable test duration. Some feasible methods to inipletnent the approach are discussed based on real data.

show abstract

Analysis of software halts in the tandem GUARDIAN operating system

Cited by 14 publications

References 9 publications

Performance Implications of Failures in Large-Scale Cluster Scheduling

Performance Implications of Failures in Large-Scale Cluster Scheduling

Measurement-Based Analysis of System Dependability Using Fault Injection and Field Failure Data

An approach to measuring and assessing dependability for critical software systems

Contact Info

Product

Resources

About