by: ali ajorian isfahan university of technology 2012 gpu architecture 1
TRANSCRIPT
![Page 1: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/1.jpg)
1
BY: ALI AJORIANISFAHAN UNIVERSITY OF TECHNOLOGY
2012
GPU Architecture
![Page 2: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/2.jpg)
2
Age of parallelism
Single CPU performance Doubled every 2 years for 30 years until 5 years ago. Marginal improvement in the last 5 years.
2005 year and checking walls Memory Wall Power Wall Processor Design Complexity
Sequential or parallel: this is the problem!!! More cores rather than more clock rate
![Page 3: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/3.jpg)
3
Early parallel computing
It was not a big idea Main frames and super computers
![Page 4: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/4.jpg)
4
And now GPUs
Stands for “Graphics Processing Unit”Integration Scheme: a card on the
motherboard with Massively Parallel computing power
![Page 5: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/5.jpg)
5
A desktop supper computer
![Page 6: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/6.jpg)
6
History of parallel computing
![Page 7: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/7.jpg)
7
GPUs: A Brief History
Stage0: graphic accelerators Early VGA cards accelerate 2D GUI Just configurable
Stage1: Fixed Graphics Hardware Graphics-only platform Very limited programmability
Stage2: GPGPU Trick GPU to do general purpose computing Programmable, but requires knowledge on computer graphics
Stream Processing Platforms High-level programming interface No knowledge on Computer Graphics is required Examples: NVIDIA’s CUDA, OpenCL
![Page 8: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/8.jpg)
8
Stream Processing Characteristics
Fairly simple computation on huge amount of data (streams) Single Program Multiple Data (SPMD)
Data Parallelism e.g., Matrix Operations, Image Processing
![Page 9: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/9.jpg)
9
Graphic accelerators to CUDA GPUs(cont)
![Page 10: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/10.jpg)
10
CUDA programming model
CPU + GPU heterogeneous programming Applications with sequential and parallel parts
Host : CPU Sequential threads
Device : GPU Parallel threads in SIMT architecture some kernels that runs on a grid of threads.
![Page 11: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/11.jpg)
11
CUDA programming model
![Page 12: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/12.jpg)
12
CUDA programming model(cont)
![Page 13: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/13.jpg)
13
GPU Architecture (NVIDIA)
![Page 14: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/14.jpg)
14
GPU Architecture (Fermi)
![Page 15: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/15.jpg)
15
SM architecture
![Page 16: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/16.jpg)
16
CUDA programming model
![Page 17: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/17.jpg)
17
Memory types
Per block registers shared memory
Per thread local memory
Per grid Global memory Constant memory Texture memory
![Page 18: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/18.jpg)
18
Memory types(cont)
![Page 19: BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e665503460f94b619ca/html5/thumbnails/19.jpg)
19
Questions?