Phoenix: Making Data-Intensive Grid Applications Fault-Tolerant

Kola, George; Kosar, Tevfik; Livny, Miron

doi:10.1109/grid.2004.51

Cited by 20 publications

(18 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A fault tolerant middleware has been described in Kola et al (2004) for data intensive distributed applications by examining the environmental conditions and classifying failures such that a suitable strategy can be applied to handle operations in a transient way. The methodology described in Kola et al (2004) uses log information from the job scheduler and the data placement scheduler and takes the next action according to user policies.…”

Section: Related Workmentioning

confidence: 99%

Error detection and error classification: failure awareness in data transfer scheduling

Balman

Kosar

2010

IJAC

View full text Add to dashboard Cite

Data transfer in distributed environment is prone to frequent failures resulting from back-end system level problems, like connectivity failure which is technically untraceable by users. Error messages are not logged efficiently, and sometimes are not relevant/useful from users' point-of-view. Our study explores the possibility of an efficient error detection and reporting system for such environments. Prior knowledge about the environment and awareness of the actual reason behind a failure would enable higher level planners to make better and accurate decisions. It is necessary to have well defined error detection and error reporting methods to increase the usability and serviceability of existing data transfer protocols and data management systems. We investigate the applicability of early error detection and error classification techniques and propose an error reporting framework and a failure-aware data transfer life cycle to improve arrangement of data transfer operations and to enhance decision making of data transfer schedulers.

show abstract

Section: Related Workmentioning

confidence: 99%

Error detection and error classification: failure awareness in data transfer scheduling

Balman

Kosar

2010

IJAC

View full text Add to dashboard Cite

show abstract

“…Phoenix [13] detects failures by scanning scheduler log files. It can diagnose execution and data transfer errors.…”

Section: Related Workmentioning

confidence: 99%

“…Hence, users can resubmit their jobs to other resources. As demonstrated in this example, both detection and recovery schemes must be an integral part of the Grid computing infrastructure [13,18]. Therefore, it is important to conduct thorough study and evaluation of new reliability models, error detection and recovery techniques before they can be deployed in production Grid environments.…”

Section: Introductionmentioning

confidence: 99%

Extending GridSim with an architecture for failure detection

Caminero

Sulistio

Caminero

et al. 2007

2007 International Conference on Parallel and Distributed Systems

View full text Add to dashboard Cite

show abstract

“…Here, each layer handles the errors within its own scope, pass others up the hierarchy [23]. This approach simplifies higher-level layers, however, at the expense of performance penalties as higher layers themselves can not adapt any more [18]. Building FT into each component in a grid is also not practical, due to scale and diversity of components.…”

Section: Failures In Grid Environmentsmentioning

confidence: 99%

“…Phoenix is detecting failures by scanning scheduler log files [18]. It can diagnose execution and data transfer errors.…”

Section: Related Workmentioning

confidence: 99%

A Service for Reliable Execution of Grid Applications

Krepska

Kielmann

Sirvent

et al.

Achievements in European Research on Grid Systems

View full text Add to dashboard Cite

In grid environments, with the large number of components (both hardware and software) that are involved in application execution, the overall probability that at least one of these components is (temporarily) non-functional is increasing rapidly. In traditional operating systems, such failures are flagged as fatal and the application will be stopped, relying on a re-start after the problem will have been fixed. In a large grid system, this is not a feasible approach as failures happen too frequently while error diagnostics might not be possible at all. This scenario is asking for a different approach to application execution, where detection and circumvention of error conditions become an integral part. We present a service that is keeping track of an application's life cycle, from submission by the user to successful completion of its execution. In a case study, we describe how GRID superscalar, a grid application programming environment, can benefit from our service.

show abstract

Phoenix: Making Data-Intensive Grid Applications Fault-Tolerant

Cited by 20 publications

References 18 publications

Error detection and error classification: failure awareness in data transfer scheduling

Error detection and error classification: failure awareness in data transfer scheduling

Extending GridSim with an architecture for failure detection

A Service for Reliable Execution of Grid Applications

Contact Info

Product

Resources

About