Adaptive checkpointing in dynamic grids for uncertain job durations

Chtepen, Maria; Dhoedt, Bart; Turck, Filip De; Demeester, Piet; Claeys, Filip; Vanrolleghem, Peter A.

doi:10.1109/iti.2009.5196152

Cited by 4 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[Zhang and Chakrabarty 2003] adaptam a frequência de checkpointing em sistemas embarcados de tempo real brando, minimizando o consumo de energia enquanto tolera um número fixo de faltas para uma dada tarefa. [Chtepen et al 2009] propõem o uso de checkpointing adaptativo em grades computacionais, ajustando o intervalo entre checkpoints de acordo com o tempo estimado de execuc ¸ão do job e da frequência de falhas de recursos; o objetivo é balancear o overhead de checkpointing e o custo de reiniciar jobs interrompidos por falhas. Em comum, esses trabalhos têm como foco controlar o tempo de execuc ¸ão limitando o tempo de recuperac ¸ão após uma falha, sem se preocupar diretamente com a responsividade.…”

Section: Trabalhos Relacionadosunclassified

Replicação de Máquinas Virtuais Xen com Checkpointing Adaptável

Silva

Obelheiro²,

Koslovki³

2015

Anais Do XVI Workshop De Testes E Tolerância a Falhas (WTF 2015)

View full text Add to dashboard Cite

Remus é um mecanismo de replicação de máquinas virtuais (MVs) que fornece alta disponibilidade diante de faltas de parada. A replicação é realizada através de checkpointing, seguindo um intervalo fixo de tempo predeterminado. Todavia, existe um antagonismo entre processamento e comunicação em relação ao intervalo ideal entre checkpoints: enquanto intervalos maiores beneficiam aplicações com processamento intensivo, intervalos menores favorecem as aplicações cujo desempenho é dominado pela rede. Logo, o intervalo utilizado nem sempre e o adequado para as características de uso de recursos da aplicação em execução na MV, limitando a aplicabilidade de Remus em determinados cenários. Este trabalho apresenta uma proposta de checkpointing adaptativo para Remus, ajustando dinamicamente a frequência de replicação de acordo com as características das aplicações em execução. Os resultados indicam que a proposta obtém um melhor desempenho de aplicações que utilizam tanto recursos de processamento como de comunicação, sem prejudicar aplicações que usam apenas um dos tipos de recursos.

show abstract

Section: Trabalhos Relacionadosunclassified

Replicação de Máquinas Virtuais Xen com Checkpointing Adaptável

Silva

Obelheiro²,

Koslovki³

2015

Anais Do XVI Workshop De Testes E Tolerância a Falhas (WTF 2015)

View full text Add to dashboard Cite

show abstract

“…Although the good results that this approaches provides, they stay limited to a static assignment. Various research is examined in this issue (Falzon & Li, 2010;Nakechbandi, Colin, & Gashumba, 2007;Chen, Zhang, & Hao 2008).Large and non dedicated computing platforms as grids may require dynamic task assignment methods to adapt to the run-time changes such as increases in the workload or resources, processor failures, and link failures (Uçar, Aykanat, Kaya, & Ikinci, 2006;He & Zhao, 2008;Chtepen, Dhoedt, Turck, & Demeester, 2008).…”

Section: Related Workmentioning

confidence: 99%

Dynamic Dependent Tasks Assignment for Grid Computing

Meddeber

Yagoubi

2011

International Journal of Grid and High Performance Computing

View full text Add to dashboard Cite

A computational grid is a widespread computing environment that provides huge computational power for large-scale distributed applications. One of the most important issues in such an environment is resource management. Task assignment as a part of resource management has a considerable effect on the grid middleware performance. In grid computing, task execution time is dependent on the machine to which it is assigned, and task precedence constraints are represented by a directed acyclic graph. This paper proposes a hybrid assignment strategy of dependent tasks in Grids which integrate static and dynamic assignment technologies. Grid computing is considered a set of clusters formed by a set of computing elements and a cluster manager. The main objective is to arrive at a method of task assignment that could achieve minimum response time and reduce the transfer cost, inducing by the tasks transfer respecting the dependency constraints.

show abstract

“…This algorithm is designed to modify a job checkpointing interval as a function of mean failure frequency of resources where the job is being executed, and the total job execution time. In [7], they developed the MeanFailureCP+ algorithm which is a modification of the MeanFailureCP that deals with checkpointing of grid applications with execution times that are unknown a priori. Review of literature reveals that a large number of research efforts have already been devoted to tolerate faults in computational grids.…”

Section: Introductionmentioning

confidence: 99%

A job checkpointing system for computational grids

Amoon¹

2013

Open Computer Science

View full text Add to dashboard Cite

Fault tolerance is an important property in computational grids since the resources are geographically distributed.Job checkpointing is one of the most common utilized techniques for providing fault tolerance in computational grids. The efficiency of checkpointing depends on the choice of the checkpoint interval. Inappropriate checkpointing interval can delay job execution. In this paper, a fault-tolerant scheduling system based on checkpointing technique is presented and evaluated. When scheduling a job, the system uses both average failure time and failure rate of grid resources combined with resources response time to generate scheduling decisions. The system uses the failure rate of the assigned resources to calculate the checkpoint interval for each job. Extensive simulation experiments are conducted to quantify the performance of the proposed system. Experiments have shown that the proposed system can considerably improve throughput, turnaround time, grid load and failure tendency of computational grids.

show abstract

Adaptive checkpointing in dynamic grids for uncertain job durations

Cited by 4 publications

References 9 publications

Replicação de Máquinas Virtuais Xen com Checkpointing Adaptável

Replicação de Máquinas Virtuais Xen com Checkpointing Adaptável

Dynamic Dependent Tasks Assignment for Grid Computing

A job checkpointing system for computational grids

Contact Info

Product

Resources

About