rsnt_translations
56,563
edits
CUDA tutorial: Difference between revisions
no edit summary
No edit summary |
No edit summary |
||
| Line 19: | Line 19: | ||
=What is a GPU?= | =What is a GPU?= | ||
A GPU, or graphics processing unit, is a single-chip processor that performs rapid mathematical calculations for the purpose of rendering images. | A GPU, or graphics processing unit, is a single-chip processor that performs rapid mathematical calculations for the purpose of rendering images. | ||
In the recent years however, such capability has been harnessed more broadly to accelerate computational workloads in cutting-edge scientific research. | |||
=What is CUDA?= <!--T:23--> | =What is CUDA?= <!--T:23--> | ||
CUDA stands for ''compute unified device architecture''. It is a scalable parallel programming model and software environment for parallel computing which | CUDA stands for ''compute unified device architecture''. It is a scalable parallel programming model and software environment for parallel computing which provides access to instructions and memory of massively parallel elements in GPUs. | ||
provides access to instructions and memory of massively parallel elements in GPUs. | |||
= | =GPU Architecture = <!--T:24--> | ||
There two main components of the GPU: | There two main components of the GPU: | ||
* Global memory | * Global memory | ||
** Similar to CPU memory | ** Similar to CPU memory | ||
** Accessible by both | ** Accessible by both CPUs and GPUs | ||
*Streaming multiprocessors (SMs) | *Streaming multiprocessors (SMs) | ||
** Each SM consists or many streaming processors (SPs) | ** Each SM consists or many streaming processors (SPs) | ||
**They perform actual computations | **They perform actual computations | ||
**Each SM has its own control | **Each SM has its own control unit, registers, execution pipelines, etc. | ||
= | |||
=Programming model= | |||
Before we start talking about the programming model, let's go over some useful terminology: | Before we start talking about the programming model, let's go over some useful terminology: | ||
*Host – The CPU and its memory (host memory) | *Host – The CPU and its memory (host memory) | ||