Efficient and Scalable Routing Algorithms for Collective Communication Operations on 2D All-Port Torus Networks

İmre, Kayhan M.; Baransel, Cesur; Artuner, Harun

doi:10.1007/s10766-011-0169-2

Cited by 12 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For different kinds of problems, different parallel algorithms have been defined that can be executed simultaneously on multiple nodes. The third section is the summation of the resulting subdivision matrices for the recursive multiplication (lines [13][14][15][16]. The pseudo-code of the algorithm is shown in Figure 1 .…”

Section: Preliminaries and Problem Statementmentioning

confidence: 99%

“…For this, we use the following equation [16]: As stated before in Section 2, alternatives will be selected based on the values for the speedup and efficiency metrics.…”

Section: Select Feasible Logical Configurationsmentioning

confidence: 99%

“…The second section is the recursive multiplication of each subsection (lines [5][6][7][8][9][10][11][12]. The third section is the summation of the resulting subdivision matrices for the recursive multiplication (lines [13][14][15][16].…”

Section: Preliminaries and Problem Statementmentioning

confidence: 99%

See 2 more Smart Citations

Systematic approach for deriving feasible mappings of parallel algorithms to parallel computing platforms

Arkin

Tekinerdoğan

İmre

2016

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

The need for high-performance computing together with the increasing trend from single processor to parallel computer architectures has leveraged the adoption of parallel computing. To benefit from parallel computing power, usually parallel algorithms are defined that can be mapped and executed on parallel computing platforms. In general, different alternative mappings can be defined each with different performance. For small computing platforms with a limited number of processing nodes, the mapping process can be carried out manually. However, for large-scale parallel computing platforms in which hundreds of thousands of processing nodes are applied, the number of possible mapping alternatives increases dramatically, and the mapping process becomes intractable for the human engineer. To assist the parallel computing engineer, we provide a systematic approach to derive feasible mapping alternatives of parallel algorithms to parallel computing platforms. The approach includes activities for modeling the parallel algorithm and parallel computing platform, generation of feasible mapping alternatives, generation of the deployment code, and finally the deployment of the generated code to the nodes. We evaluate our approach for deriving feasible mapping alternatives for four well-known parallel algorithms. The evaluation is based on both simulations and real executions of the generated mapping alternatives.

show abstract

Section: Preliminaries and Problem Statementmentioning

confidence: 99%

“…For this, we use the following equation [16]: As stated before in Section 2, alternatives will be selected based on the values for the speedup and efficiency metrics.…”

Section: Select Feasible Logical Configurationsmentioning

confidence: 99%

See 1 more Smart Citation

Systematic approach for deriving feasible mappings of parallel algorithms to parallel computing platforms

Arkin

Tekinerdoğan

İmre

2016

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Fig. 1 shows the complete exchange algorithm which purpose is to collect all data from all nodes and to distribute data to all nodes [9][10] [11]. This algorithm is a commonly used parallel algorithm that is often used as part of a bigger parallel algorithm.…”

Section: Problem Statementmentioning

confidence: 99%

Model-Driven Approach for Supporting the Mapping of Parallel Algorithms to Parallel Computing Platforms

Arkin

Tekinerdoğan

İmre

2013

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. The trend from single processor to parallel computer architectures has increased the importance of parallel computing. To support parallel computing it is important to map parallel algorithms to a computing platform that consists of multiple parallel processing nodes. In general different alternative mappings can be defined that perform differently with respect to the quality requirements for power consumption, efficiency and memory usage. The mapping process can be carried out manually for platforms with a limited number of processing nodes. However, for exascale computing in which hundreds of thousands of processing nodes are applied, the mapping process soon becomes intractable. To assist the parallel computing engineer we provide a model-driven approach to analyze, model, and select feasible mappings. We describe the developed toolset that implements the corresponding approach together with the required metamodels and model transformations. We illustrate our approach for the well-known complete exchange algorithm in parallel computing.

show abstract

“…An effective multicast routing should minimize communication latency which consists of startup latency, network latency, and blocking latency [9]. The startup latency is the amount of time incurred by the operating system in preparing the message for injection into the network.…”

Section: Introductionmentioning

confidence: 99%

A tree-based algorithm for multicasting in 2D torus networks

El-Baky

2015

Egyptian Informatics Journal

View full text Add to dashboard Cite

Torus network has become increasingly important to multicomputer design because of its many desirable properties including low bandwidth and fixed degree of nodes. Also, torus networks can be partitioned into mesh networks. The multicast pattern, in which one source node sends the same message to multiple destination nodes, is the essential pattern in a wide variety of applications. This paper proposes a multicast routing scheme in 2D torus networks. The proposed scheme is a Tree-based Algorithm which Splits torus Networks into two Equally Meshes, hence it is called TASNEM. TASNEM algorithm is a tree-based technique, in which the router simultaneously sends incoming flits on more than one outgoing channel. It requires at most two start-up times, one for each mesh subnetwork. For each mesh subnetwork, the source node delivers a message to the destination nodes along one main path and different horizontal paths branched from the main path. TASNEM algorithm can achieve high degree of parallelism and low communication latency over a wide range of traffic loads. Performance results of a simulation study on torus networks are discussed to compare TASNEM algorithm with some previous algorithms.Ó 2015 Production and hosting by Elsevier B.V. on behalf

show abstract

Efficient and Scalable Routing Algorithms for Collective Communication Operations on 2D All-Port Torus Networks

Cited by 12 publications

References 17 publications

Systematic approach for deriving feasible mappings of parallel algorithms to parallel computing platforms

Systematic approach for deriving feasible mappings of parallel algorithms to parallel computing platforms

Model-Driven Approach for Supporting the Mapping of Parallel Algorithms to Parallel Computing Platforms

A tree-based algorithm for multicasting in 2D torus networks

Contact Info

Product

Resources

About