[CUDA study notes] First acquaintance with CUDA

---

Foreword

This article is my study notes for the CUDA course, mainly for myself to review the past and learn new things in the future.It's an honor to be of any help to you.If you are a beginner in CUDA, please correct me if I am wrong.If there is any infringement, please contact the author to delete it.

---

I. What is GPU?

Graphics Processing Unit (English: Graphics Processing Unit, abbreviation: GPU; also known as display core, graphics card, visual processor, display chip or graphics chip) is a specialized computer, workstation, game console and some mobile devices.Microprocessors (such as tablets, smartphones, etc.) that perform drawing operations.

Second, what is CUDA?

CUDA (Compute Unified Device Architecture, unified computing architecture) is an integrated technology launched by NVIDIA, which is the company's solution to GPGPU (English: General-purpose computing on graphics processing units, referred to as GPGPU or GP²U))official name.Through this technology, users can use NVIDIA's GPU to perform operations other than image processing. It is also the first time that a GPU can be used as a development environment for a C-compiler.The CUDA development kit (CUDA Toolkit) can only compile its own CUDA C-language (only the function of linking to OpenCL), that is, the part executed on the GPU is compiled into the PTX intermediate language or the machine code of a specific NVIDIA GPU architecture (NVIDIA official nameIt is "device code"); and the C/C++ code (officially called "host code" by NVIDIA) that executes on the central processing unit still depends on an external compiler, such as Microsoft Visual Studio under Microsoft Windows; under Linux, it is mainlyDepends on GCC.

Three. Graphics card structure diagram

Below is a schematic of a newer graphics card (TESLA V1000).The green ones are all cores, with a total of more than 5,000 cores.

Fourth, CPU vs GPU

1. GPU can be dozens of times faster than CPU: cpu: general computer has 4-8 cpu cores, desktop computers can reach 24; GPU: can reach 5, 6 thousand.runs very fast,
2. GPUs have great potential for scientific computing.
   

V. Cooperation between CPU and GPU:

1. The GPU is connected to the CPU through a high-speed interface: PCIe
2. GPU as acceleration component.
   

Six, GPU application: