An analysis of clustered failures on large supercomputing systems

Hacker, Thomas J.; Romero, Fabián; Carothers, Christopher D.

doi:10.1016/j.jpdc.2009.03.007

Cited by 55 publications

(63 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to influence risk assessment along any business/third party grounds the management and the setting of the risk assessment is taken outside of the cloud fabric and turned it into a service level [4]. In terms of wider resource failure a wide range of studies exist for distributed computing environments [32], [33]. Risk from a third party service as an extension to risk assessment mechanisms has also been explored in cloud environments [34].…”

Section: Related Workmentioning

confidence: 99%

A Risk Assessment Framework for Cloud Computing

Djemame

Armstrong

Guitart

et al. 2016

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

Abstract-Cloud service providers offer access to their resources through formal Service Level Agreements (SLA), and need wellbalanced infrastructures so that they can maximise the Quality of Service (QoS) they offer and minimise the number of SLA violations. This paper focuses on a specific aspect of risk assesment as applied in cloud computing: methods within a framework that can be used by cloud service providers and service consumers to assess risk during service deployment and operation. It describes the various stages in the service lifecycle wheres risk assessment takes place, and the corresponding risk models that have been designed and implemented. The impact of risk on architectural components, with special emphasis on holistic management support at service operation, is also described. The risk assessor is shown to be effective through the experimental evaluation of the implementation, and is already integrated in a cloud computing toolkit.

show abstract

Section: Related Workmentioning

confidence: 99%

A Risk Assessment Framework for Cloud Computing

Djemame

Armstrong

Guitart

et al. 2016

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

show abstract

“…The key parameter is the MTBF µ = 1 λ . Weibull distributions are a good example of probability distributions that account for infant mortality, and they are widely used to model failures on computer platforms [42,67,54,39,43]. The definition of Weibull(λ ), the Weibull distribution law of shape parameter k and scale parameter λ , goes as follows:…”

Section: Resilience At Scalementioning

confidence: 99%

“…But if k < 1, the failure rate decreases with time, and the smaller k, the more important the decreasing. Values used in the literature are k = 0.7 or k = 0.5 [39,54,67].…”

Section: Resilience At Scalementioning

confidence: 99%

Fault-Tolerance Techniques for High-Performance Computing

2015

Computer Communications and Networks

View full text Add to dashboard Cite

This report provides an introduction to resilience methods. The emphasis is on checkpointing, the de-facto standard technique for resilience in High Performance Computing. We present the main two protocols, namely coordinated checkpointing and hierarchical checkpointing. Then we introduce performance models and use them to assess the performance of theses protocols. We cover the Young/Daly formula for the optimal period and much more! Next we explain how the efficiency of checkpointing can be improved via fault prediction or replication. Then we move to application-specific methods, such as ABFT. We conclude the report by discussing techniques to cope with silent errors (or silent data corruption).This report is a slightly modified version of the first chapter of the monograph Fault tolerance techniques for high-performance computing edited by Thomas Herault and Yves Robert, and to be published by Springer Verlag.

show abstract

“…A number of studies have looked at resource failures in distributed environments [28,8,29,30,31,32,33,34,35]. Schroeder and Gibson [28] analyse failure data collected over 9 years at Los Alamos National Laboratory (LANL), and includes 23,000 failures recorded on more than 20 different systemsmostly large clusters of Symmetric-Multi-Processing (SMP) and Non-Uniform-Memory-Access (NUMA) nodes.…”

Section: Related Workmentioning

confidence: 99%

Resource failures risk assessment modelling in distributed environments

Alsoghayer

Djemame

2014

Journal of Systems and Software

View full text Add to dashboard Cite

Service providers offer access to resources and services in distributed environments such as Grids and Clouds through formal Service level Agreements (SLA), and need well-balanced infrastructures so that they can maximise the Quality of Service (QoS) they offer and minimise the number of SLA violations. We propose a mathematical model to predict the risk of failure of resources in such environments using a discrete-time analytical model driven by reliability functions fitted to observed data. The model relies on the resource historical data so as to predict the risk of failure for a given time interval. The model is evaluated by comparing the predicted risk of failure with the observed risk of failure, and is shown to accurately predict the resources risk of failure, allowing a service provider to selectively choose which SLA request to accept.

show abstract

An analysis of clustered failures on large supercomputing systems

Cited by 55 publications

References 5 publications

A Risk Assessment Framework for Cloud Computing

A Risk Assessment Framework for Cloud Computing

Fault-Tolerance Techniques for High-Performance Computing

Resource failures risk assessment modelling in distributed environments

Contact Info

Product

Resources

About