From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

CPU vs. GPU: Architectural difference

CPU vs. GPU: Architectural difference - NVIDIA Tutorial

From the course: NVIDIA Certified Associate AI Infrastructure and Operations (NCA-AIIO) Cert Prep

Start my 1-month free trial Buy for my team

CPU vs. GPU: Architectural difference

“

Let me give you a brief idea about high-level architecture of CPUs versus GPU. I'm not sure how many of you have this understanding, so I'm going to give you a very basic information about different architectural component of a CPU. So inside a CPU, there would be cores, like here it is a core, here it is a core, here it is a core, here it is a core. So this is your quad-core CPU, there are four cores. And then there are cache. Here it is L1 cache, level 1 cache. Then there could be L2 cache. And there can be L3 cache. This is a common architecture for modern CPUs where they have multiple cores. And they have their ALU, as we call it, arithmetic logical unit or control unit, as you can refer to them. And that allows you to perform control mechanism. And all of them may have their own individual cache, some common caches and maybe another larger L3 cache, level 3 cache. This is how a CPU architecture looks like. So like if I'm talking about Intel i9, i9 has 24 cores in it, right. Now…

Contents