Scheduling heuristics for efficient broadcast operations on grid environments

Barchet-Steffenel,; Mounie,

doi:10.1109/ipdps.2006.1639634

Cited by 8 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the computational grid becomes a widely accept paradigm for large-scale parallel systems, collective communications have had a lot of attention recently. Angelo et al [1] focused on the development of scheduling techniques to minimize makespan of a broadcast operation on a grid environment. Different to the above focus, Beaumont et al [3][4][5] concentrated upon the problem of broadcasting in heterogeneous platforms with a one port model.…”

Section: Related Workmentioning

confidence: 99%

“…Communications over HN are a point-to-point blocking send operation. 1 3. Each processor within HN has unique ID.…”

Section: Preliminariesmentioning

confidence: 99%

“…Grid computing, the most referred paradigm of heterogeneous networks, integrates geographically distributed computing resources to establish a virtual and high expandable parallel machine [1]. It presents a new trend to distribute and internet computing for coordinating the sharing of large-scale heterogeneous resources and for solving problems in dynamic, multi-institutional virtual organizations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Scheduling for atomic broadcast operation in heterogeneous networks with one port model

Hsu

Tsai

2009

J Supercomput

View full text Add to dashboard Cite

With the emergence of the network technologies, heterogeneous computing has become a wide accept paradigm for distributed and network computing. In this paper, we present different algorithms aiming to efficiently perform atomic oneto-all broadcast in a heterogeneous network with a one port model. The proposed algorithms are divided into graph-based and tree-based ones. In graph-based algorithms, we present Nearest Neighbor First and Maximum Degree Neighbor First schemes. A prescheduling strategy with constructing a message forwarding table for avoiding redundant transmissions is applied as runtime support. In the tree-based approaches, there are five heuristic algorithms: Nearest Neighbor First, Maximum Degree Neighbor First, Maximum Height Subtree First, Maximum Subtree First, and Maximum Weighted Subtree First, proposed based on different network characteristics. To evaluate the performance of the proposed techniques, we have developed a simulator that contains a parametric graph generator for generating network graphs with various characteristics. We have implemented all of the proposed scheduling algorithms on the simulator. The performance results show that the Maximum Weighted Subtree First performs best in high degree heterogeneous environments. On the contrary, with homogeneous-like environments, the graph-based Nearest Neighbor First will be the best choice. In summary, contribution of this study relies on informing significant suggestions for adapting proper broadcasting mechanism in different heterogeneous platforms.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Communications over HN are a point-to-point blocking send operation. 1 3. Each processor within HN has unique ID.…”

Section: Preliminariesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Scheduling for atomic broadcast operation in heterogeneous networks with one port model

Hsu

Tsai

2009

J Supercomput

View full text Add to dashboard Cite

show abstract

“…Using inter-cluster communication parameters, we can construct an optimized broadcast tree between clusters using scheduling heuristics, an approach that provides better performances on grid environments than traditional grid-unaware algorithms found on most MPI distributions [3]. Indeed, in this example we rely on the Early Completion Edge First -ECEF heuristic, proposed by [4].…”

Section: Second Hierarchical Levelmentioning

confidence: 99%

“…We can eventually integrate the communication time inside each cluster, as stated by the ECEF-LAt heuristic [3], where the communication schedule minimizes the overall broadcast time by taking into account both inter and intracluster communication times…”

Section: Second Hierarchical Levelmentioning

confidence: 99%

Adaptive Performance Modeling on Hierarchical Grid Computing Environments

Nasri

Steffenel²,

Trystram³

2007

Seventh IEEE International Symposium on Cluster Computing and the Grid (CCGrid '07)

View full text Add to dashboard Cite

show abstract

Assessing Contention Effects on MPI_Alltoall Communications

Steffenel

Martinasso

Trystram

2007

Advances in Grid and Pervasive Computing

View full text Add to dashboard Cite

One of the most important collective communication patterns used in scientific applications is the complete exchange, also called All-to-All. Although efficient algorithms have been studied for specific networks, general solutions like those available in well-known MPI distributions (e.g. the MPI_Alltoall operation) are strongly influenced by the congestion of network resources. In this paper we present an integrated approach to model the performance of the All-to-All collective operation, which consists in identifying a contention signature that characterizes a given network environment, using it to augment a contention-free communication model. This approach, assessed by experimental results, allows an accurate prediction of the performance of the All-to-All operation over different network architectures with a small overhead.

show abstract

Scheduling heuristics for efficient broadcast operations on grid environments

Abstract: Abstract

Cited by 8 publications

References 16 publications

Scheduling for atomic broadcast operation in heterogeneous networks with one port model

Scheduling for atomic broadcast operation in heterogeneous networks with one port model

Adaptive Performance Modeling on Hierarchical Grid Computing Environments

Assessing Contention Effects on MPI_Alltoall Communications

Contact Info

Product

Resources

About