[CS] GPU

1. What is a GPU?

- Acronym for Graphics Processing Units

- Specialized processor originally designed to accelerate graphics rendering

* Rendering: the process of generating images

- Able to process many pieces of data simultaneously (=parallel processing)

- made up of many smaller and more specialized cores

- used in a wide range of applications (e.g., graphics, video rendering, gaming, AI, HPC, etc)

*HPC: stands for High Performance Computing / the ability to process data and perform complex calculations at high speeds

 

2. GPU vs CPU

- Both are silicon-based microprocessors

 

-CPU-

- suited to a wide variety of tasks, especially those for which latency or per-core performance are important (e.g., web browsing)

- it focuses its smaller number of cores on individual tasks and getting things done quickly

 

-GPU-

- designed for a specific purpose

- Over time, it became more flexible

- the GPU evolved as a complement to the CPU

 

3. GPU vs Graphics Card

While they are often used interchangeably, there is a subtle distinction between the two.

Much like a motherboard contains a CPU, a graphics card refers to an add-in board that incorporates the GPU.

 

4. GPU's 2 basic types

1) integrated

the majority of GPUs on the market are integrated type

embedded alongside the CPU

does not come on its own separate card at all

A CPU that comes with a fully integrated GPU allows for thinner and lighter systems, reduced power consumption, and lower system costs.

 

2) discrete

for more resource-intensive applications with extensive performance demands, the discrete type is better than the integrated type

a distinct chip that is mounted on its own circuit board

this type adds processing power at the cost of additional energy consumption and heat creation

it generally requires dedicated cooling for maximum performance

 

Source:

https://www.intel.com/content/www/us/en/products/docs/processors/what-is-a-gpu.html