Generating Parallel Algorithms for Cluster and Grid Computing

Hayashida, Ulisses Kendi; Okuda, Kunio; Panetta, Jairo; Song, Siand Wun

doi:10.1007/11428831_63

Cited by 5 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We selected a matrix multiplication BSP application, developed by Hayashida et al [22], since it is a highly coupled and long-running parallel application and produces Fig. 6 Execution overhead of the checkpointing mechanism large checkpoints.…”

Section: Storage Of Checkpoints From Parallel Applicationsmentioning

confidence: 99%

Reliable management of checkpointing and application data in opportunistic grids

2010

View full text Add to dashboard Cite

Opportunistic computational grids use idle processor cycles from shared machines to enable the execution of long-running parallel applications. Besides computational power, these applications may also consume and generate large amounts of data, requiring an efficient data storage and management infrastructure. In this article, we present an integrated middleware infrastructure that enables the use of not only idle processor cycles, but also unused disk space of shared machines. Our middleware enables the reliable distributed storage of application data in the shared machines in a redundant and fault-tolerant way. A checkpointing-based mechanism monitors the execution of parallel applications, saves periodical checkpoints in the shared machines, and in case of node failures, supports the application migration across heterogeneous grid nodes. We evaluate the feasibility of our middleware using experiments and simulations. Our evaluation shows that the proposed middleware promotes important improvements in grid data management reliability while imposing a low performance overhead.

show abstract

Section: Storage Of Checkpoints From Parallel Applicationsmentioning

confidence: 99%

Reliable management of checkpointing and application data in opportunistic grids

2010

View full text Add to dashboard Cite

show abstract

“…? In [2] we use the systolic array structure to solve two basic problems: matrix product and alignment of two strings. We now use the matrix product example to illustrate the redundancy method.…”

Section: Matrix Multiplication Examplementioning

confidence: 99%

“…However, as observed in [2], the basic systolic algorithm is not suitable for cluster computing because of the fine granularity and the large number of processors required. To make the granularity coarser we consider sub-matrices instead of single elements in the basic algorithm.…”

Section: Matrix Multiplication Examplementioning

confidence: 99%

See 1 more Smart Citation

Reliable Systolic Computing Through Redundancy

Okuda

Song

Yamamoto

2006

Advances in Computer Systems Architecture

Self Cite

View full text Add to dashboard Cite

Abstract. The systolic array paradigm has low communication demand because it does not use costly global communication and each processor communicates with few other processors. It is thus suitable to be used in cluster computing. The systolic approach, however, is vulnerable in a heterogeneous environment where machines perform differently. In this paper we propose a redundant systolic solution with high-availability to deal with this problem. We analyze the overhead that results from the need to coordinate the actions of the redundant processors and show that this overhead is worth the performance improvement it provides.

show abstract

“…Alves et al [13] present a systolic algorithm to compare two strings and obtain a similarity measure. Hayashida et al [28] attempt to generalize the usage of the systolic paradigm to design parallel algorithms for cluster and grid computing. Their approach is based on a modification of the dependence transformation method, originally proposed to design systolic arrays for VLSI implementation on silicon chips [29,30].…”

mentioning

confidence: 99%

Performance results of running parallel applications on the InteGrade

Cáceres

Mongelli

Loureiro

et al. 2010

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe InteGrade middleware intends to exploit the idle time of computing resources in computer laboratories. In this work we investigate the performance of running parallel applications with communication among processors on the InteGrade grid. As costly communication on a grid can be prohibitive, we explore the so-called systolic or wavefront paradigm to design the parallel algorithms in which no global communication is used. To evaluate the InteGrade middleware we considered three parallel algorithms that solve the matrix chain product problem, the 0-1 Knapsack Problem, and the local sequence alignment problem, respectively. We show that these three applications running under the InteGrade middleware and MPI take slightly more time than the same applications running on a cluster with only LAM-MPI support. The results can be considered promising and the time difference between the two is not substantial. The overhead of the InteGrade middleware is acceptable, in view of the benefits obtained to facilitate the use of grid computing by the user. These benefits include job submission, checkpointing, security, job migration, etc.

show abstract

Generating Parallel Algorithms for Cluster and Grid Computing

Cited by 5 publications

References 6 publications

Reliable management of checkpointing and application data in opportunistic grids

Reliable management of checkpointing and application data in opportunistic grids

Reliable Systolic Computing Through Redundancy

Performance results of running parallel applications on the InteGrade

Contact Info

Product

Resources

About