Checkpointing-based rollback recovery for parallel applications on the InteGrade grid middleware

Camargo, Raphael Y. de; Goldchleger, Andrei; Kon, Fábio; Goldman, Alfredo

doi:10.1145/1028493.1028499

Cited by 21 publications

(14 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The adoption of a checkpoint-based approach introduces, independently of the occurrence of failures, an overhead to the normal application execution time. Integrade checkpointing implementation minimizes this overhead by copying the checkpoint data to a buffer and performing the coding and transfer of checkpoints through a separate application thread, allowing the application to concurrently continue its execution [3].…”

Section: Integrade Fault Tolerance Mechanismmentioning

confidence: 99%

See 1 more Smart Citation

A Flexible Fault-Tolerance Mechanism for the Integrade Grid Middleware

Sousa¹,

Silva²,

Lopes

2007

International Conference on Networking and Services (ICNS '07)

View full text Add to dashboard Cite

Abstract-Computer grids have attracted great attention of both academic and enterprise communities, becoming an attractive alternative for the execution of applications that demand huge computational power, allowing the integration of computational resources spread through different administrative domains. The dynamic nature of the grid infrastructure, its high scalability, and great heterogeneity exacerbates the likelihood of errors occurrence, imposing fault tolerance as a major requirement for grid middlewares. This paper describes a flexible fault-tolerance mechanism implemented on Integrade grid middleware that allows the customization of several failure handling parameters and the combination of different failure handling techniques. This paper also presents several experiments that measure the benefits of our approach, considering several different execution environments scenarios. I. INTRODUCTIONA computer grid comprises a hardware and software infrastructure that allows integration and sharing of distributed resources, such as software, data and peripherals, inside and among institutions. This computational infrastructure has attracted great attention of academic and enterprise communities, becoming an attractive alternative for execution of applications that demand huge computational power, and allowing the integration of computational resources spread through different administrative domains.Computational grids have been used to solve problems in varied areas of scientific, enterprise, and industrial activities, such as: computational biology, image processing for medical diagnosis, weather forecast, high energy physics, marketing simulations, and oil prospection. Grid computing has empower the conception of a new generation of applications that allow combining computations, experiments, observations, and data got in real time. The phenomena modeled by these applications require diverse software components whose compositions and interactions are extremely dynamic. Moreover, the grid infrastructure is also heterogeneous and dynamic, aggregating a great amount of computation and communication resources, databases and, sometimes, sensors and specific peripherals. The dynamism can be observed in terms of high variation in resource availability, node instability, and workload variations in nodes and network links.The dynamic nature of the grid infrastructure, its high scalability, and great heterogeneity has turn impracticable its

show abstract

Section: Integrade Fault Tolerance Mechanismmentioning

confidence: 99%

“…Checkpoint data recovery also involves a query to the CDRM, requesting the list of ADRs where the application checkpoints were stored. More details about Integrade checkpointing mechanism can be found in [2], [3]. Figure 2 illustrates our implementation of the Integrade protocol for executing application replicas.…”

Section: Integrade Fault Tolerance Mechanismmentioning

confidence: 99%

A Flexible Fault-Tolerance Mechanism for the Integrade Grid Middleware

Sousa¹,

Silva²,

Lopes

2007

International Conference on Networking and Services (ICNS '07)

View full text Add to dashboard Cite

show abstract

“…There is now a rich literature on checkpointing techniques for parallel computation on a cluster [1,7,25,26,28,34,35]. Nevertheless, a thorny issue remains.…”

Section: Introductionmentioning

confidence: 99%

Transparent adaptive library-based checkpointing for master-worker style parallelism

Cooperman

Ansel

2006

Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06)

View full text Add to dashboard Cite

show abstract

“…They provide a C API that allows applications written in both C and C++ to use them. The basic checkpointing functionality is provided by functions to manipulate the checkpoint stack, to save the stack data to a file, and to recover checkpointing data [28].…”

Section: The Integrade Grid Middlewarementioning

confidence: 99%

Checkpointing of Parallel Applications in a Grid Environment

Sajadah

Terstyansky

Winter

et al.

Distributed and Parallel Systems

View full text Add to dashboard Cite

This is an electronic version of an MPhil thesis awarded by the University of Westminster.This is an exact reproduction of the paper copy held by the University of Westminster library.The WestminsterResearch online digital archive at the University of Westminster aims to make the research output of the University available to a wider audience. Copyright and Moral Rights remain with the authors and/or copyright owners. Users are permitted to download and/or print one copy for non-commercial private study or research. Further distribution and any use of material from within this archive for profit-making enterprises or for commercial gain is strictly forbidden.Whilst further distribution of specific materials from within this archive is forbidden, you may freely distribute the URL of WestminsterResearch: (http://westminsterresearch.wmin.ac.uk/). Many thanks are due to all those who read this document and spent hours helping me amassing the information used here.And finally, I wish to thank all my family and beloved friends for all these singular years.Abstract ii AbstractThe Grid environment is generic, heterogeneous, and dynamic with lots of unreliable resources making it very exposed to failures. The environment is unreliable because it is geographically dispersed involving multiple autonomous administrative domains and it is composed of a large number of components. Examples of failures in the Grid environment can be: application crash, Grid node crash, network failures, and Grid system component failures. These types of failures can affect the execution of parallel/distributed application in the Grid environment and so, protections against these faults are crucial. Therefore, it is essential to develop efficient fault tolerant mechanisms to allow users to successfully execute Grid applications. One of the research challenges in Grid computing is to be able to develop a fault tolerant solution that will ensure Grid applications are executed reliably with minimum overhead incurred.While checkpointing is the most common method to achieve fault tolerance, there is still a lot of work to be done to improve the efficiency of the mechanism. This thesis provides an in-depth description of a novel solution for checkpointing parallel applications executed on a Grid. The checkpointing mechanism implemented allows to checkpoint an application at regions where there is no interprocess communication involved and therefore reducing the checkpointing overhead and checkpoint size.

show abstract

Checkpointing-based rollback recovery for parallel applications on the InteGrade grid middleware

Cited by 21 publications

References 15 publications

A Flexible Fault-Tolerance Mechanism for the Integrade Grid Middleware

A Flexible Fault-Tolerance Mechanism for the Integrade Grid Middleware

Transparent adaptive library-based checkpointing for master-worker style parallelism

Checkpointing of Parallel Applications in a Grid Environment

Contact Info

Product

Resources

About