<aside> ❗ Headings and text with “▶” can be toggled to expand contained info. You can also see the Table of Contents as a collapsible sidebar on the right →

</aside>

Table of Contents

Introduction

This module is dedicated to GPU microarchitecture concepts. Students will familiarize themselves with essential terminologies and crucial architecture principles pertaining to GPUs. Through this learning, they will acquire a comprehensive understanding of GPU microarchitecture and its significance.

Objectives

• Describe GPU microarchitecture
• Be able to explain the basic GPU architecture terminologies

Readings

Required

Optional

Notes / Things that stand out

Lesson 1 - Multi-Threaded Architecture

Recap: GPU Architecture

• Lesson 4 - Introduction to GPU Architecture

Multi-Threading

Figure 1: GPU with 5 stage pipeline and 4 instructions

Ideal Case

Cache Miss (in-order processor)

• In an in-order processor, if an instruction has a cache miss, then the following instructions have to wait until the first instruction completes.
• However, Instructions 2 and 3 don’t depend on Instruction 1 (see Figure 1), only Instruction 4 does.