Improving availability with recursive microreboots: a soft-state system case study

Candea, George; Cutler, James; Fox, Armando

doi:10.1016/j.peva.2003.07.007

Cited by 71 publications

(46 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All runtime data will be collected. [4] Result analysis: After acquiring reliability evaluation results, system operators, application developers and middleware vendors can study the result carefully to find out the best-of-the-breed fault tolerant configuration, select another middleware product or redesign some parts of the system. The second and the third steps are iterative.…”

Section: Approach Overviewmentioning

confidence: 99%

“…In Java, for example, "when a program violates the semantic constraints of the Java programming language, the Java virtual machine signals this error to the program as an exception" [24]. Some experiments in [4] show that almost all underlying faults can be manifested by Java exceptions. At the same time, in terms of the philosophy of exception handling, throwing and catching exceptions become the most popular mechanisms to deal with faults when programming middleware products and applications.…”

Section: Fault Modelmentioning

confidence: 99%

See 1 more Smart Citation

Simulation-based analysis of middleware service impact on system reliability: Experiment on Java application server

Huang

Wang

Liu

et al. 2011

Journal of Systems and Software

View full text Add to dashboard Cite

Section: Approach Overviewmentioning

confidence: 99%

Section: Fault Modelmentioning

confidence: 99%

Simulation-based analysis of middleware service impact on system reliability: Experiment on Java application server

Huang

Wang

Liu

et al. 2011

Journal of Systems and Software

View full text Add to dashboard Cite

“…The possibility of returning to the initial state at any time is common in reactive systems to prevent resource leaks in long-running executions, for instance through micro-reboots [2]. Communication protocols in the IP and telephony communities regulate the coexistence of interacting features.…”

Section: Well-architected Systemsmentioning

confidence: 99%

Extending Symmetry Reduction by Exploiting System Architecture

Trefler

Wahl

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Symmetry reduction is a technique to alleviate state explosion in model checking by replacing a model of replicated processes with a bisimilar quotient model. The size of the quotient depends strongly on the set of applicable symmetries, which in many practical cases allows only polynomial reduction. We introduce architectural symmetry, a concept that exploits architectural system features to compensate for a lack of symmetry in the system model. We show that the standard symmetry quotient of an architecturally symmetric and well-architected model preserves arbitrary Boolean combinations and nestings of reachability properties. This quotient can be exponentially smaller than the model, even in cases where traditional symmetry reduction is nearly ineffective. Our technique thus extends the benefits of symmetry reduction to systems that are in fact not symmetric. Finally, we generalize our results to all architecturally symmetric models, including those that are not well-architected. We illustrate our method through examples and experimental data.

show abstract

“…In this section, we discuss our application of recursive restarts (RR), 41 a recently proposed technique for achieving high availability that exploits partial restarts at various levels within complex software infrastructures to recover from transient failures and rejuvenate software components such that overall mean-time-to-recover (MTTR) is minimized. 6 We had two main goals in applying RR to Mercury. The first was to partially remove the human from the loop in ground station control by automating recovery from common transient failures that were restart curable.…”

Section: B High Availability Upgradesmentioning

confidence: 99%

“…GSML has enabled independent development of commodity ground station services and a framework for extending these services to commodity satellite operations tasks. 22,31 Our application of recursive restarts, 6 a ROC high-availability technique, has reduced recovery time by a factor of four in Mercury systems and eliminated the adverse effects of many ground station software failure modes. Mercury support of a scientific small satellite mission, QuakeSat-1, 14,24 has provided low-cost flexible satellite contact opportunities that currently do not exist among commercial providers.…”

Section: Introductionmentioning

confidence: 99%

A Framework For Robust and Flexible Ground Station Networks

Cutler

Fox

2006

Journal of Aerospace Computing, Information, and Communication

Self Cite

View full text Add to dashboard Cite

Space communication systems are plagued by complexity resulting in limited access to orbiting satellites and high mission operation costs that ultimately reduce mission yields and capabilities. In particular, ground stations, the access point between space and terrestrial networks, suffer from monolithic designs, narrow interfaces, and reliability issues that raise significant engineering barriers. We have reduced these barriers by decomposing monolithic ground station functions into basic component services and virtualizing their software and hardware interfaces. We captured these core services in a command and control protocol, called the Ground Station Markup Language. This modularization has also enabled us to apply recovery-oriented computing techniques, specifically recursive restarts, to quickly recover from software failures and improve overall ground station availability. Our testbed consists of globally distributed ground stations supporting university satellite missions. With case studies, we demonstrate how our work enables rapid prototyping and reconfiguration within a ground station and how it facilitates coordination of networked, heterogeneous ground stations by hiding low-level implementation details.

show abstract

Improving availability with recursive microreboots: a soft-state system case study

Cited by 71 publications

References 50 publications

Simulation-based analysis of middleware service impact on system reliability: Experiment on Java application server

Simulation-based analysis of middleware service impact on system reliability: Experiment on Java application server

Extending Symmetry Reduction by Exploiting System Architecture

A Framework For Robust and Flexible Ground Station Networks

Contact Info

Product

Resources

About