Efficient NAS Benchmark Kernels with C++ Parallel Programming

Griebler, Dalvan; Löff, Júnior; Mencagli, Gabriele; Danelutto, Marco; Fernandes, Luiz Gustavo

doi:10.1109/pdp2018.2018.00120

Cited by 33 publications

(31 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…-4 sequential NAS kernels (IS, FT, EP and MG), C++ version [11] as reference; finite difference 3D heat resolution (7-point stencil) ( [15] as reference); -6D Vlasov-Poisson, using Fortran/MPI Selalib [16] as reference.…”

Section: Discussionmentioning

confidence: 99%

InKS: a programming model to decouple algorithm from optimization in HPC codes

Ejjaaouani

Aumage²,

Bigot

et al. 2019

J Supercomput

View full text Add to dashboard Cite

Existing programming models tend to tightly interleave algorithm and optimization in HPC simulation codes. This requires scientists to become experts in both the simulated domain and the optimization process and makes the code difficult to maintain or port to new architectures. In this paper, we propose the InKS programming model that decouples these concerns with two distinct languages: InKS pia to express the simulation algorithm and InKS pso for optimizations. We define InKS pia and evaluate the feasibility of defining InKS pso with three test-languages: InKS o/C++ , InKS o/loop , InKS o/XMP . We evaluate the approach on synthetic benchmarks (NAS and heat equation) as well as on a more complex example (6D Vlasov-Poisson solver). Our evaluation demonstrates the soundness of the approach as it improves the separation of algorithmic and optimization concerns at no performance cost. We also identify a set of guidelines for the later full definition of the InKS pso language.

show abstract

Section: Discussionmentioning

confidence: 99%

InKS: a programming model to decouple algorithm from optimization in HPC codes

Ejjaaouani

Aumage²,

Bigot

et al. 2019

J Supercomput

View full text Add to dashboard Cite

show abstract

“…Over the last decade, various studies have parallelized the Barnes-Hut algorithm for shared memory as well as for distributed memory architectures. However, the algorithm has mostly been parallelized using the native languages of HPC such as UPC [21] and Fortran [22]. Various researchers have contended that Java could be deemed as a potential language for high-performance computing.…”

Section: Methodsologymentioning

confidence: 99%

On the parallelization and performance analysis of Barnes–Hut algorithm using Java parallel platforms

et al. 2020

View full text Add to dashboard Cite

Multi-core processors provide time-efficient and cost-effective solutions to execute the algorithms for complex physical systems. However, to efficiently exploit the processing capabilities of the underlying architectures, the applications executing on multi-core processors must be parallelized. Conventionally, the applications for high-performance computing (HPC) are written in native (programming) languages. However, several researchers have argued that Java can be an excellent alternative for writing HPC applications. To gauge the comparative performance of Java parallel platforms, a benchmark study is required considering some real HPC applications. In this study, as a pioneer application, Java-based parallel platforms, namely Java thread API, JavaSymphony, and MPJ Express, are used to parallelize and benchmark the Barnes-Hut algorithm to simulate the N-body physical system of celestial objects. The parallel implementations of the Barnes-Hut algorithm are tested for performance on shared memory multi-core architectures. The hardware-level performance analysis involving the parameters like the number of instructions executed by the CPU, level-1 and level-3 cache misses, and the number of main memory accesses is conducted to investigate the insights of the attained performances. The celestial objects up to one million are simulated, and the results revealed that the JavaSymphony-based implementation of the Barnes-Hut algorithm produces better results as compared to the other employed parallel frameworks.

show abstract

“…In this section, we investigate the frameworks for the performance of their transformed codes. We use two popular benchmark suites: PolyBench and NAS parallel benchmarks (NPB) . PolyBench is a suite of linear algebra kernels targeted for auto‐parallelizers based on the polyhedral model.…”

Section: Quantitative Study Of Auto‐parallelization Frameworkmentioning

confidence: 99%

A study on popular auto‐parallelization frameworks

Prema

Nasre

Jehadeesan

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

We study five popular auto-parallelization frameworks (Cetus, Par4all, Rose, ICC, and Pluto) andcompare them qualitatively as well as quantitatively. All the frameworks primarily deal with loop parallelization but differ in the techniques used to identify parallelization opportunities. Due to this variance, various aspects, such as certain loop transformations, are supported only in a few frameworks. The frameworks exhibit varying abilities in handling loop-carried dependence and, therefore, achieve different amounts of speedup on widely used PolyBench and NAS parallel benchmarks. In particular, Intel C Compiler (ICC) fares as an overall good parallelizer. Our study also highlights the need for more sophisticated analyses, user-driven parallelization, and meta-auto-parallelizer that provides combined benefits of various frameworks. KEYWORDS loop-carried dependence, loop transformations, privatization, vectorization INTRODUCTIONQuest for performance has made multicore processors mainstream. It becomes vital for the programmers and algorithm developers to exploit these ubiquitous advanced architectures with parallel programming. Exploiting the potential of multicore processors through parallel programming is a significant challenge. Amid several approaches to tame this challenge, a promising and programmer-friendly approach is automatic parallelization. 1-3 Auto-parallelizers eliminate the need for a programmer to transform a sequential code into a parallel code, which is quite attractive.Source-to-source transformations with insertion of parallel directives are performed by auto-parallelizers, such as Cetus, 4-7 Par4all, 8,9 Pluto,10,11 Parallware, 12,13 Rose, 14,15 Intel C Compiler, 16 LLVM (Low Level Virtual Machine) Polly, 17 ParaWise, 18 ParaGraph, 19 SUIF (Stanford University Intermediate Format), 20-22 and Polaris. 23,24 Although these existing parallelizers offer considerable benefits, they still fall short of fully replacing the manual transformations. Several parallelizers do not utilize all the static information available, whereas several others fall short of modeling precision. It leads to either missed parallelism opportunities or unwanted parallelization of sequential codes. As a consequence, compared to the original sequential version, the auto-parallelized code may lead to parallelization overheads and may exhibit poorer performance. 25Earlier studies 26,27 of the parallelizing compilers using Perfect benchmarks performed a detailed analysis of the code restructuring techniques.The techniques include induction variable elimination, scalar expansion, forward substitution, strip mining, and loop interchange. The studies found that some of the programs showed improvements and that scalar expansion led to positive results. An effectiveness study on the Polaris compiler with OpenMP (Open Multi-Processing) 28 parallel code using Perfect benchmarks proclaimed a performance lag in small parallel loops. It also illustrated the importance of reduction operation, which resulted in a moderate (10%) performance...

show abstract

Efficient NAS Benchmark Kernels with C++ Parallel Programming

Cited by 33 publications

References 13 publications

InKS: a programming model to decouple algorithm from optimization in HPC codes

InKS: a programming model to decouple algorithm from optimization in HPC codes

On the parallelization and performance analysis of Barnes–Hut algorithm using Java parallel platforms

A study on popular auto‐parallelization frameworks

Contact Info

Product

Resources

About