IB Computer Science HL · Year 2 · Lesson 37

CPU/GPU, pipelining + A1 Paper 1 synthesis

Computer performance emerges from the interaction of processing, memory, storage and specialized hardware. Data representation and Boolean logic connect physical binary states to numbers, media, decisions and digital circuits.

Paper 150 minutesA1.1.2 · A1.1.3 · A1.1.6 · A1.1.1 · A1.1.4 · A1.1.5 · A1.1.7…
Use Next or the arrow keys.
Course homeIA hub2027 case study
Today’s targets

What you need to be able to do

2 / 9
  • DescribeThe role of a GPU
  • ExplainThe differences between the CPU and the GPU
  • DescribeThe process of pipelining in multi-core architectures
  • RetrieveReconnect today’s new material to previously taught content from this topic.
A1.1.2A1.1.3A1.1.6A1.1.1A1.1.4A1.1.5A1.1.7A1.1.8A1.1.9A1.2.1A1.2.2A1.2.3A1.2.4A1.2.5A1.3.1A1.3.2A1.3.3A1.3.4A1.3.5A1.3.6
Paper 1 lensMatch the depth of every response to the command term. Previously learned content can move quickly, but retrieval must still be accurate.
Page 2 of 9
Retrieve

Rapid Recall Deck

3 / 9

Say the answer aloud before flipping. Mark secure knowledge quickly and spend time on the gaps.

  • Why are GPUs effective for highly parallel workloads?: They contain many processing units designed to perform similar operations across large amounts of data concurrently.
  • How do CPU and GPU design priorities differ?: CPUs emphasize flexible, low-latency execution of varied tasks; GPUs emphasize high-throughput parallel processing.
  • How can CPUs and GPUs work together?: The CPU coordinates general program logic and transfers suitable parallel work/data to the GPU, then combines or uses the results.
  • What is instruction pipelining?: Overlapping stages of different instructions so one instruction can be fetched while others are being decoded, executed, or written back.
  • What are the main instruction-pipeline stages?: Fetch, decode, execute, and write-back.
  • How do multiple CPU cores improve throughput?: Independent cores can execute different instruction streams/tasks in parallel when the workload can be divided.
  • How do the ALU, control unit, registers, and buses work together during instruction processing?: The control unit coordinates execution, registers hold immediate instruction/data state, buses transfer addresses/data/control signals, and the ALU performs required arithmetic or logical operations.
Page 3 of 9
A1.1.2 + A1.1.3 · Learn

Core knowledge and application

4 / 9
A1.1.2Describe

The role of a GPU

Computer performance emerges from the interaction of processing, memory, storage and specialized hardware.

Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1The architecture that allows graphics processing units (GPUs) to handle specific tasks and makes them suitable for complex computations
Required detail 2Apply to video games, artificial intelligence (AI), large simulations and other applications that require graphics rendering and machine learning

Explain it without notes

Describe: The role of a GPU in the context of a school upgrading computers for graphics, coding and machine learning.

  • The architecture that allows graphics processing units (GPUs) to handle specific tasks and makes them suitable for complex computations
  • Apply to video games, artificial intelligence (AI), large simulations and other applications that require graphics rendering and machine learning
A1.1.3Explain

The differences between the CPU and the GPU

Computer performance emerges from the interaction of processing, memory, storage and specialized hardware.

Exam moveGive the mechanism or relationship and make the reasons/causes explicit.
Required detail 1Differences in their design philosophies, usage scenarios
Required detail 2Differences in their core architecture, processing power, memory access, power efficiency
Required detail 3CPUs and GPUs working together: task division, data sharing, coordinating execution

Explain it without notes

Explain: The differences between the CPU and the GPU in the context of a school upgrading computers for graphics, coding and machine learning.

  • Differences in their design philosophies, usage scenarios
  • Differences in their core architecture, processing power, memory access, power efficiency
  • CPUs and GPUs working together: task division, data sharing, coordinating execution
Page 4 of 9
A1.1.6 · Learn

Core knowledge and application

5 / 9
A1.1.6Describe

The process of pipelining in multi-core architectures

Computer performance emerges from the interaction of processing, memory, storage and specialized hardware.

Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1The instructions fetch, decode, execute
Required detail 2Write-back stages to improve the overall system performance in multi-core architectures
Required detail 3Understand how cores in multi-core processors work independently and in parallel

Explain it without notes

Describe: The process of pipelining in multi-core architectures in the context of a school upgrading computers for graphics, coding and machine learning.

  • The instructions fetch, decode, execute
  • Write-back stages to improve the overall system performance in multi-core architectures
  • Understand how cores in multi-core processors work independently and in parallel
Page 5 of 9
2027 Case Study connection

CPU/GPU architecture and diffusion-model workload

6 / 9

CPU strength

General-purpose cores are optimized for flexible control-heavy workloads and low-latency execution of varied tasks.

GPU strength

Many processing units can execute large amounts of similar numerical work in parallel, which suits matrix/tensor operations common in neural-network inference and training.

Case-study challengeUse architecture, parallelism, memory access, power/throughput and CPU–GPU task division to explain why iterative denoising can be computationally expensive and how hardware choices affect deployment.
Evidence bankAdd one sourced example comparing generative-model performance or resource use across hardware/deployment options.
Page 6 of 9
Apply

Transfer to a new scenario

7 / 9
ScenarioA school upgrading computers for graphics, coding and machine learning needs a design or technical decision related to today’s topic. Explain what matters and why.
  • The architecture that allows graphics processing units (GPUs) to handle specific tasks and makes them suitable for complex computations
  • Apply to video games, artificial intelligence (AI), large simulations and other applications that require graphics rendering and machine learning
  • Differences in their design philosophies, usage scenarios
  • Differences in their core architecture, processing power, memory access, power efficiency
  • The instructions fetch, decode, execute
  • Write-back stages to improve the overall system performance in multi-core architectures
Page 7 of 9
Exam lens

Paper 1 practice

8 / 9
Build the response before checking notesUse precise terminology and match the required depth.
  1. Describe: The role of a GPU in the context of a school upgrading computers for graphics, coding and machine learning.
  2. Explain: The differences between the CPU and the GPU in the context of a school upgrading computers for graphics, coding and machine learning.
  3. Describe: The process of pipelining in multi-core architectures in the context of a school upgrading computers for graphics, coding and machine learning.
Self-checkAnswer the exact command term. For explain, include mechanism/reason; for compare, pair criteria; for discuss/evaluate/justify, build supported reasoning and a conclusion.
Page 8 of 9
Homework

Finish the learning cycle

9 / 9

IA — main task

Finish Criterion D evidence: development documentation, testing evidence, video planning/recording and full source-code appendix.

Syllabus — short

Complete a short Paper 1 retrieval set based on today’s topic.

Designed by David Xu
© 2026 David Xu. All rights reserved.
Educators and students are welcome to use these materials for non-commercial teaching and learning with attribution. Please share the original link when possible. Reposting, redistributing modified copies, removing attribution, or commercial use requires prior permission.
End of Lesson 37.