Data Streams, Data-flow parallelism, and Real-time Analytics [email protected] -- EPFL DATA Lab.
Parallelism Real Time
-
Upload
elton-vinson -
Category
Documents
-
view
28 -
download
3
description
Transcript of Parallelism Real Time
![Page 1: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/1.jpg)
Parallelism Real TimeParallelism Real Time
Anca BatoriLavinia
BasarabăAnca
Brandimbur
“Politehnica” University of TimişoaraAutomation and Computing University
Computers Department
iunie 2010
![Page 2: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/2.jpg)
ObjectivesObjectives
presenting the use of graphical processing units (GPU) to achieve significant improvements for real time systems
presenting the main steps for developing an application using CUDA
offering a source of resources
![Page 3: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/3.jpg)
CConontentstentsIntroductionParallelism Application descriptionExperimentsConclusions
![Page 4: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/4.jpg)
IntroductionIntroductionReal Time Systems are an
important area of research and development
Many applications: Airplanes, biomedical accelerators, nuclear power plants
Necessity of parallelism to achieve desired time limits
![Page 5: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/5.jpg)
ParallelismParallelismCan be hardware and software
◦GPUs represent a combinationGPUs have certain characteristcs,
that CPUs do not poses, that can be useful for certain application◦More processing power, less
flexibilityThe application are developed
using CUBLAS
![Page 6: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/6.jpg)
AArrcchitectuhitecturesresCPU CPU vsvs GPU GPU
Cores number 4Threads 2Cache memory
=> random address acces
Cores number 240Threads 1024Cache memory =>
fast access to consecutive addresses
Sursa: www.nvidia.com
![Page 7: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/7.jpg)
Support Vector Machines Support Vector Machines (SVM)(SVM)
Possibilities to split two classes
![Page 8: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/8.jpg)
SVMSVM geometrical geometrical representationrepresentation
![Page 9: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/9.jpg)
CPU versus GPUCPU versus GPUSpeedup 54x for1000 imagesEven better results for bigger training
sets
![Page 10: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/10.jpg)
Multiclass Multiclass SVMSVM
“One Against All” (OAA)training: M binary classifiers(M number of
classes)testing: strategy„the winner takes it all”
“One Against One” (OAO)training : M(M-1)/2 binary classifiers testing : strategy maximum number of votes
“Directed Acyclic Graph” (DAG)testing : decision tree
![Page 11: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/11.jpg)
Multiclass AlgorithmsMulticlass Algorithms
NVIDIA GTX 280IDSIA sets
Training set dimension
Training time (secunde) Recognition percentage (%)
OAO DAG OAA OAO DAG OAA
100 15,183 15,183 7,092 84,94 84,45 67,39
300 129,188 129,188 77,438 88,37 87,91 82,19
1000 1806 1806 1452,7 90,73 89,84 82,94
![Page 12: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/12.jpg)
Multiclass Multiclass AlgoritAlgorithhmmss
![Page 13: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/13.jpg)
ConclusionsConclusionsIn the period of ubiquitous and
pervasive systems, real time systems are a very important field
Since CPUs do not evolve, so fast as they used to, there is a shift towards parallel programming and more and more systems are developed this way
GPUs can improve the execution time for an application 100x times
![Page 14: Parallelism Real Time](https://reader030.fdocuments.in/reader030/viewer/2022032612/568131f7550346895d9852d5/html5/thumbnails/14.jpg)
Thank youThank you