Application of Multiblock Grid and Dual-Level Parallelism in Coastal Ocean Circulation Modeling

Luong, Phu; Breshears, Clay P.; Ly, Le Ngoc

doi:10.1023/b:jomp.0000008722.81924.50

Cited by 3 publications

(8 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other authors rely on libraries, such as Cowles (2008) Aggregating the simulation results of each submesh does not yield the same result as simulating the original full mesh, in other words, completely separate calculations for each submesh are not possible. As shown in Figure 3 every slave also needs to consult and compute points around its submesh, known as ghost points (Luong et al, 2004;Henshawa and Schwendemanb, 2008;Tseng and Chien, 2011) orhalo points (Sannino et al, 2001;Cowles, 2008;Wang et al, 2010). This implies not only more redundant computationin the slaves, as can be seen in the following algorithm description, but also the need to exchange some data between neighbouring slaves through the master at every time step.…”

Section: Calculate the Vertical Velocity Component Equation (12)mentioning

confidence: 99%

“…Most of the ocean models developed have been parallelized (Sannino et al, 2001; Luong et al, 2004; Fringer et al, 2006; Cowles, 2008; Henshawa and Schwendemanb, 2008; Wang et al, 2010; Tseng and Chien, 2011). Each of these has its own characteristics such as the type of mesh used, how the mesh is divided, how the different submeshes overlap, how the workload is distributed between the processing units, if they are models for distributed or shared memory computers, and the programming model used.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient parallelization of a regional ocean model for the western Mediterranean Sea

Córdoba

García-Dopico

García

et al. 2013

The International Journal of High Performance Computing Applica

View full text Add to dashboard Cite

This paper focuses on the parallelization of an ocean model applying current multicore processor-based cluster architectures to an irregular computational mesh. The aim is to maximize the efficiency of the computational resources used. To make the best use of the resources offered by these architectures, this parallelization has been addressed at all the hardware levels of modern supercomputers: firstly, exploiting the internal parallelism of the CPU through vectorization; secondly, taking advantage of the multiple cores of each node using OpenMP; and finally, using the cluster nodes to distribute the computational mesh, using MPI for communication within the nodes. The speedup obtained with each parallelization technique as well as the combined overall speedup have been measured for the western Mediterranean Sea for different cluster configurations, achieving a speedup factor of 73.3 using 256 processors. The results also show the efficiency achieved in the different cluster nodes and the advantages obtained by combining OpenMP and MPI versus using only OpenMP or MPI. Finally, the scalability of the model has been analysed by examining computation and communication times as well as the communication and synchronization overhead due to parallelization.

show abstract

Section: Calculate the Vertical Velocity Component Equation (12)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Efficient parallelization of a regional ocean model for the western Mediterranean Sea

Córdoba

García-Dopico

García

et al. 2013

The International Journal of High Performance Computing Applica

View full text Add to dashboard Cite

show abstract

“…A parallel version of MGPOM uses MPI asynchronous sends and receives to exchange data between adjacent blocks at the interfaces. OpenMP has been used as a second level of parallelization within each MPI process to improve the load balance in the simulation of the Arabian Gulf [7]. This area is extended from 48 East to 58 East in longitude and from 23.5 North to 30.5 North in latitude (left part of Figure 2).…”

Section: Application Of Nanoscompiler Extensions To Mgpommentioning

confidence: 99%

“…This area is extended from 48 East to 58 East in longitude and from 23.5 North to 30.5 North in latitude (left part of Figure 2). The computational 20-block grid shown on the right part of Figure 2 (also used in [7]) is used in this study.…”

Section: Application Of Nanoscompiler Extensions To Mgpommentioning

confidence: 99%

Dual-Level Parallelism Exploitation with OpenMP in Coastal Ocean Circulation Modeling

González¹,

Ayguadé²,

Martorell³

et al. 2002

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Two alternative dual-level parallel implementations of the Multiblock Grid Princeton Ocean Model (MGPOM) are compared in this paper. The first one combines the use of two programming paradigms: message passing with the Message Passing Interface (MPI) and shared memory with OpenMP (version called MPI-OpenMP); the second uses only OpenMP (version called OpenMP-Only). MGPOM is a multiblock grid code that enables the exploitation of two levels of parallelism. The MPI-OpenMP implementation uses MPI to parallelize computations by assigning each grid block to a unique MPI process. Since not all grid blocks are of the same size, the workload between processes varies. OpenMP is used within each MPI process to improve load balance. The alternative OpenMP-Only implementation uses some extensions proposed to OpenMP that defines thread groups in order to efficiently exploit the available two levels of parallelism. These extensions are supported by a research OpenMP compiler named NanosCompiler. Performance results of the two implementations from the MGPOM code on a 20-block grid for the Arabian Gulf simulation demonstrate the efficacy of the OpenMP-Only versions of the code. The simplicity of the OpenMP implementation as well as the possibility of using and simply defining policies to dynamically change the allocation of OpenMP threads to the two levels of parallelism is the main result of this study and suggests to consider this alternative for the parallelization of future applications. Keywords: OpenMP and MPI implementations, multiple levels of parallelism, multiblock grid, coastal ocean circulation model.

show abstract

“…Duan Geng [39] used the hybrid MPI/OpenMP programming model to improve the KMP algorithm. Phu Luong [40] applied dual-Level Parallelism in coastal ocean circulation modeling. Furthermore, the hybrid parallel method is applied in many new developing fields, e.g., machine learning [41][42][43], data mining [44], and cloud computing [45].…”

Section: Introductionmentioning

confidence: 99%

Matrix Information Geometry for Signal Detection via Hybrid MPI/OpenMP

Feng

Hua

Wang

et al. 2019

Entropy

View full text Add to dashboard Cite

The matrix information geometric signal detection (MIGSD) method has achieved satisfactory performance in many contexts of signal processing. However, this method involves many matrix exponential, logarithmic, and inverse operations, which result in high computational cost and limits in analyzing the detection performance in the case of a high-dimensional matrix. To address these problems, in this paper, a high-performance computing (HPC)-based MIGSD method is proposed, which is implemented using the hybrid message passing interface (MPI) and open multiple processing (OpenMP) techniques. Specifically, the clutter data are first modeled as a Hermitian positive-definite (HPD) matrix and mapped into a high-dimensional space, which constitutes a complex Riemannian manifold. Then, the task of computing the Riemannian distance on the manifold between the sample data and the geometric mean of these HPD matrices is assigned to each MPI process or OpenMP thread. Finally, via comparison with a threshold, the signal is identified and the detection probability is calculated. Using this approach, we analyzed the effect of the matrix dimension on the detection performance. The experimental results demonstrate the following:(1) parallel computing can effectively optimize the MIGSD method, which substantially improves the practicability of the algorithm; and (2) the method achieves superior detection performance under a higher dimensional HPD matrix.Entropy 2019, 21, 1184 2 of 15 manifold theory plays in statistical information. After that, series of statistical inference theories were investigated via information geometry [15,16]. From the perspective of information geometry, P(X|θ ) is regarded as a point on a statistical manifold, with a cylindrical confidence zone R that is centered on θ 0 , where the parameter θ 0 represents the null hypothesis sample data and M denotes the statistical manifold. After the statistical modeling from the observed sample data, we can determine whether θ is equal to θ 0 or not. Figure 1 illustrates the basic principle of the statistical hypothesis problem.Entropy 2019, 21, x FOR PEER REVIEW 2 of 15 manifold theory plays in statistical information. After that, series of statistical inference theories were investigated via information geometry [15,16]. From the perspective of information geometry,

show abstract

Application of Multiblock Grid and Dual-Level Parallelism in Coastal Ocean Circulation Modeling

Cited by 3 publications

References 4 publications

Efficient parallelization of a regional ocean model for the western Mediterranean Sea

Efficient parallelization of a regional ocean model for the western Mediterranean Sea

Dual-Level Parallelism Exploitation with OpenMP in Coastal Ocean Circulation Modeling

Matrix Information Geometry for Signal Detection via Hybrid MPI/OpenMP

Contact Info

Product

Resources

About