Stability of multi-cluster synchronization

A system of many coupled oscillators on a network can show multicluster synchronization. We obtain existence conditions and stability bounds for such a multicluster synchronization. When the oscillators are identical, we obtain the interesting result that network structure alone can cause multicluster synchronization to emerge even when all the other parameters are the same. We also study occurrence of multicluster synchronization when two different types of oscillators are coupled.

Strategies for Rescheduling Tightly-Coupled Parallel Applications in Multi-Cluster Grids

As computational Grids are increasingly used for executing long running multi-phase parallel applications, it is important to develop efficient rescheduling frameworks that adapt application execution in response to resource and application dynamics. In this paper, three strategies or algorithms have been developed for deciding when and where to reschedule parallel applications that execute on multi-cluster Grids. The algorithms derive rescheduling plans that consist of potential points in application execution for rescheduling and schedules of resources for application execution between two consecutive rescheduling points. Using large number of simulations, it is shown that the rescheduling plans developed by the algorithms can lead to large decrease in application execution times when compared to executions without rescheduling on dynamic Grid resources. The rescheduling plans generated by the algorithms are also shown to be competitive when compared to the near-optimal plans generated by brute-force methods. Of the algorithms, genetic algorithm yielded the most efficient rescheduling plans with 9-12% smaller average execution times than the other algorithms.

Dynamic parallel job scheduling in multi-cluster computing systems

Job scheduling is a complex problem, yet it is fundamental to sustaining and improving the performance of parallel processing systems. In this paper, we address an on-line parallel job scheduling problem in heterogeneous multi-cluster computing systems. We propose a new space-sharing scheduling policy and show that it performs substantially better than the conventional policies.

Multi-cluster computing interconnection network performance modeling and analysis

The overall performance of a distributed system is often depends on the effectiveness of its interconnection network. Thus, the study of the communication networks for distributed systems is very important, which is the focus of this paper. In particular, we address the problem of fat-tree based interconnection networks performance modeling for multi-user heterogeneous multi-cluster computing systems. To this end, we present an analytical model and validate the model through comprehensive simulation. The results of the simulation demonstrated that the proposed model exhibits a good degree of accuracy for various system organizations and under different working conditions.

Analytical modeling of interconnection networks in heterogeneous multi-cluster systems

In data-intensive distributed systems, replication is the most widely used approach to offer high data availability, low bandwidth consumption, increased fault-tolerance and improved scalability of the overall system. Replication-based systems implement replica control protocols that enforce a specified semantics of accessing the data. Also, the performance depends on a number of factors, the chief of which is the protocol used to maintain consistency among object replica. In this paper, we propose a new low-cost and high data availability protocol called the box-shaped grid structure for maintaining consistency of replicated data on networked distributed computing systems. We show that the proposed protocol provides high data availability, low communication costs, and increased fault-tolerance as compared to the baseline replica control protocols.

Performance analysis of interconnection networks for multi-cluster systems

With the current popularity of cluster computing systems, it is increasingly important to understand the capabilities and potential performance of various interconnection networks. In this paper, we propose an analytical model for studying the capabilities and potential performance of interconnection networks for multi-cluster systems. The model takes into account stochastic quantities as well as network heterogeneity in bandwidth and latency in each cluster. Also, blocking and non-blocking network architecture model is proposed and are used in performance analysis of the system. The model is validated by constructing a set of simulators to simulate different types of clusters, and by comparing the modeled results with the simulated ones.

A performance model for analysis of heterogeneous multi-cluster systems

This paper addresses the problem of performance modeling for large-scale heterogeneous distributed systems with emphases on multi-cluster computing systems. Since the overall performance of distributed systems is often depends on the effectiveness of its communication network, the study of the interconnection networks for these systems is very important. Performance modeling is required to avoid poorly chosen components and architectures as well as discovering a serious shortfall during system testing just prior to deployment time. However, the multiplicity of components and associated complexity make performance analysis of distributed computing systems a challenging task. To this end, we present an analytical performance model for the interconnection networks of heterogeneous multi-cluster systems. The analysis is based on a parametric family of fat-trees, the m-port n-tree, and a deterministic routing algorithm, which is proposed in this paper. The model is validated through comprehensive simulation, which demonstrated that the proposed model exhibits a good degree of accuracy for various system organizations and under different working conditions.

Performance analysis of heterogeneous multi-cluster systems

When building a cost-effective high-performance parallel processing system, a performance model is a useful tool for exploring the design space and examining various parameters. However, performance analysis in such systems has proven to be a challenging task that requires the innovative performance analysis tools and methods to keep up with the rapid evolution and ever increasing complexity of such systems. To this end, we propose an analytical model for heterogeneous multi-cluster systems. The model takes into account stochastic quantities as well as network heterogeneity in bandwidth and latency in each cluster. Also, blocking and non-blocking network architecture model is proposed and are used in performance analysis of the system. The message latency is used as the primary performance metric. The model is validated by constructing a set of simulators to simulate different types of clusters, and by comparing the modeled results with the simulated ones.

Analysis of interconnection networks in heterogeneous multi-cluster systems

The study of interconnection networks is important because the overall performance of a distributed system is often critically hinged on the effectiveness of its interconnection network. In the mean time, the heterogeneity is one of the most important factors of such systems. This paper addresses the problem of interconnection networks performance modeling of large-scale distributed systems with emphases on heterogeneous multi-cluster computing systems. So, we present an analytical model to predict message latency in multi-cluster systems in the presence of cluster size heterogeneity. The model is validated through comprehensive simulation, which demonstrates that the proposed model exhibits a good degree of accuracy for various system organizations and under different working conditions.

Analytical interconnection networks model for multi-cluster computing systems

This paper addresses the problem of interconnection networks performance modeling of large-scale distributed systems with emphases on multi-cluster computing systems. The study of interconnection networks is important because the overall performance of a distributed system is often critically hinged on the effectiveness of its interconnection network. We present an analytical model that considers stochastic quantities as well as processor heterogeneity of the target system. The model is validated through comprehensive simulation, which demonstrates that the proposed model exhibits a good degree of accuracy for various system sizes and under different operating conditions.

Analysis of multi-cluster computing systems with processor heterogeneity

This paper addresses the problem of performance modeling of heterogeneous multi-cluster computing systems. We present an analytical model that can be employed to explore the effectiveness of different design approaches so that one can have an intelligent choice during design and evaluation of a cost effective large-scale heterogeneous distributed computing system. The proposed model considers stochastic quantities as well as processor heterogeneity of the target system. The analysis is based on a parametric fat-tree network, the m-port n-tree, and a deterministic routing algorithm. The correctness of the proposed model is validated through comprehensive simulation of different types of clusters.

Analytical modeling of communication latency in multi-cluster systems

This paper addresses the problem of performance modeling of large-scale distributed systems with emphasis on communication networks in heterogeneous multi-cluster systems. The study of interconnection networks is important because the overall performance of a distributed system is often critically hinged on the effectiveness of this part. We present an analytical model to predict message latency in multi-cluster systems in the presence of processor heterogeneity. The model is validated through comprehensive simulation, which demonstrates that the proposed model exhibits a good degree of accuracy for various system sizes and under different operating conditions.

Multi-cluster computing interconnection network performance modeling and analysis

The overall performance of a distributed system often depends on the effectiveness of its interconnection network. Thus, the study of the communication networks for distributed systems–which is the focus of this paper–is very important. In particular, we address the problem of fat-tree based interconnection networks performance modeling for multi-user heterogeneous multi-cluster computing systems. To this end, we present an analytical model and validate the model through comprehensive simulation. The results of the simulation demonstrate that the proposed model exhibits a good degree of accuracy for various system organizations and under different working conditions. On the basis of the validated model, we propose an adaptive assignment function based on the existing heterogeneity of the system to minimize multi-user environment overhead.

Visualização simultânea e multi-nível de informações de monitoramento de cluster

Clusters de computadores são geralmente utilizados para se obter alto desempenho na execução de aplicações paralelas. Sua utilização tem aumentado significativamente ao longo dos anos e resulta hoje em uma presença de quase 60% entre as 500 máquinas mais rápidas do mundo. Embora a utilização de clusters seja bastante difundida, a tarefa de monitoramento de recursos dessas máquinas é considerada complexa. Essa complexidade advém do fato de existirem diferentes configurações de software e hardware que podem ser caracterizadas como cluster. Diferentes configurações acabam por fazer com que o administrador de um cluster necessite de mais de uma ferramenta de monitoramento para conseguir obter informações suficientes para uma tomada de decisão acerca de eventuais problemas que possam estar acontecendo no seu cluster. Outra situação que demonstra a complexidade da tarefa de monitoramento acontece quando o desenvolvedor de aplicações paralelas necessita de informações relativas ao ambiente de execução da sua aplicação para entender melhor o seu comportamento. A execução de aplicações paralelas em ambientes multi-cluster e grid juntamente com a necessidade de informações externas à aplicação é outra situação que necessita da tarefa de monitoramento. Em todas essas situações, verifica-se a existência de múltiplas fontes de dados independentes e que podem ter informações relacionadas ou complementares. O objetivo deste trabalho é propor um modelo de integração de dados que pode se adaptar a diferentes fontes de informação e gerar como resultado informações integradas que sejam passíveis de uma visualização conjunta por alguma ferramenta. Esse modelo é baseado na depuração offline de aplicações paralelas e é dividido em duas etapas: a coleta de dados e uma posterior integração das informações. Um protótipo baseado nesse modelo de integração é descrito neste trabalho Esse protótipo utiliza como fontes de informação as ferramentas de monitoramento de cluster Ganglia e Performance Co-Pilot, bibliotecas de rastreamento de aplicações DECK e MPI e uma instrumentação do Sistema operacional Linux para registrar as trocas de contexto de um conjunto de processos. Pajé é a ferramenta escolhida para a visualização integrada das informações. Os resultados do processo de integração de dados pelo protótipo apresentado neste trabalho são caracterizados em três tipos: depuração de aplicações DECK, depuração de aplicações MPI e monitoramento de cluster. Ao final do texto, são delineadas algumas conclusões e contribuições desse trabalho, assim como algumas sugestões de trabalhos futuros.