Multi approach for real-time systems specification: case study of GPU parallel systems

Zouaneb, Imane; Belarbi, Mostefa; Chouarfia, Abdellah

doi:10.1504/ijbdi.2016.077385

Cited by 13 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Authors in [5] proposes a multi approach for real-time systems specification and design. The purpose of the work is to couple MARTE [4] with the Event-B method.…”

Section: Related Workmentioning

confidence: 99%

A Formal Approach for the Modeling of High-Level Architectures Aligned with System Requirements

Bougacha

2020

Rigorous State-Based Methods

View full text Add to dashboard Cite

“…Authors in [5] proposes a multi approach for real-time systems specification and design. The purpose of the work is to couple MARTE [4] with the Event-B method.…”

Section: Related Workmentioning

confidence: 99%

A Formal Approach for the Modeling of High-Level Architectures Aligned with System Requirements

Bougacha

2020

Rigorous State-Based Methods

View full text Add to dashboard Cite

“…Theneedforfasterexecutionhasbecomeamatterofforemostimportanceasdatasetskeepgrowing biggerandtheexecutionofcomplexoperationswiththesedatasetscantakealotoftimetorun.GPU computinghasprovidedacomprehensivesolutiontothisproblem (Zouaneb,Belarbi,&Chouarfia, 2016).AGraphicProcessingUnitisamany-coremultithreadedmultiprocessorwhichusesmassively parallel computing techniques for achieving higher efficiency and superior performance. It is a methodofcomputationwhicheffectivelyreducestheworkloadoftheCPU.AGPUhashundreds ofcoreascomparedtothatofaCPUwhichhasamaximumoffouroreightcores.Thisenables ThefollowingarefamousformatsthatareimprovisedversionsoftheCSRformat:…”

Section: Gpu Computingmentioning

confidence: 99%

GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format

Mohammed

Babu

Sriram

2018

International Journal of Grid and High Performance Computing

View full text Add to dashboard Cite

With the growth of data parallel computing, role of GPU computing in non-graphic applications such as image processing becomes a focus in research fields. Convolution is an integral operation in filtering, smoothing and edge detection. In this article, the process of convolution is realized as a sparse linear system and is solved using Sparse Matrix Vector Multiplication (SpMV). The Compressed Sparse Row (CSR) format of SPMV shows better CPU performance compared to normal convolution. To overcome the stalling of threads for short rows in the GPU implementation of CSR SpMV, a more efficient model is proposed, which uses the Adaptive-Compressed Row Storage (A-CSR) format to implement the same. Using CSR in the convolution process achieves a 1.45x and a 1.159x increase in speed compared to the normal convolution of image smoothing and edge detection operations, respectively. An average speedup of 2.05x is achieved for image smoothing technique and 1.58x for edge detection technique in GPU platform usig adaptive CSR format.

show abstract

“…Also,findingthebest-fitsettingforbuildingalogisticmodelneedsacertainamountofcross-validation experimentsthatcanalsobeverytime-consuming (Diamantidis,Karlis,&Giakoumakis,2000). Asmanytechniqueswereproposedtospeedupthetrainingprocessandgethighperformance computation,forexampleMultithreaded,Multi-coreCPUs,MessagePassingInterface(MPI)and recently Open Computing Language (OpenCL) (Zouaneb, Belarbi, & Chouarfia, 2016). Each technologyhasitsowndeploymentcharacteristicsandexecutioncost.Nowadays,manyresearchers focus in parallelizing a variety of complex computational algorithms (Lotrič & Dobnikar, 2009) basedonrecentcapabilitiesofnewhardwarearchitectures.OpenCLisoneofHighperformance computation and development framework that is developed by Khronos group to expand GPUs computationfromnotonlygraphicsrending,butalsogeneralpurposecomputation(oftencalled GeneralPurposeGPU:GPGPU) (Harris,2005) (Witten,Frank,Hall,&Pal,2016).OpenCLprovides an industry standard for parallel programming of heterogeneous computing platforms (Zouaneb, Belarbi,&Chouarfia,2016),itisnotdedicatedtospecificGPUvendorslikeComputeUnifiedDevice Architecture(CUDA)whichisrestrictedonlyforNVidiaGPUs.TherecentdevelopmentsofGPU haveshownasuperbcomputationalperformancewiththecurrentmulti-coreCPUs.Nevertheless, GPUsarespeciallydesignedtofacilitateacceleratedgraphicsprocessing;theyhavebeenusedas generalpurposecomputingdevices.CertainhighperformanceGPUsarenowdesignedtoexecutes general-purposeprocessesinsteadofgraphicsrendering,whichitwastheonlyusageofGPUsbefore (Munshi,2009) Thispaperisorganizedasfollows.Insection2,wediscussedtherelevantworksthathavedone beforeinacceleratingthetrainingprocesswithrespecttoourapproach.Theproposedapproachis presentedindetailinsection3alongwithsequentialtrainingprocessoflogisticregression.Section 4analysestheresultsobtainedfromrunningseveralexperimentsondifferentplatforms.Finally,the paperisconcludedinsection5.…”

Section: Introductionmentioning

confidence: 99%

Accelerating Training Process in Logistic Regression Model using OpenCL Framework

Zahera

El-Sisi

2017

International Journal of Grid and High Performance Computing

View full text Add to dashboard Cite

In this paper, the authors propose a new parallel implemented approach on Graphics Processing Units (GPU) for training logistic regression model. Logistic regression has been applied in many machine learning applications to build building predictive models. However, logistic training regularly requires a long time to adapt an accurate prediction model. Researchers have worked out to reduce training time using different technologies such as multi-threading, Multi-core CPUs and Message Passing Interface (MPI). In their study, the authors consider the high computation capabilities of GPU and easy development onto Open Computing Language (OpenCL) framework to execute logistic training process. GPU and OpenCL are the best choice with low cost and high performance for scaling up logistic regression model in handling large datasets. The proposed approach was implement in OpenCL C/C++ and tested by different size datasets on two GPU platforms. The experimental results showed a significant improvement in execution time with large datasets, which is reduced inversely by the available GPU computing units.

show abstract

Multi approach for real-time systems specification: case study of GPU parallel systems

Cited by 13 publications

References 13 publications

A Formal Approach for the Modeling of High-Level Architectures Aligned with System Requirements

A Formal Approach for the Modeling of High-Level Architectures Aligned with System Requirements

GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format

Accelerating Training Process in Logistic Regression Model using OpenCL Framework

Contact Info

Product

Resources

About