Table of Contents
Programming Heterogeneous Computing Systems with GPUs and other Accelerators (227-0085-51L)
Course Description
The increasing difficulty of scaling the performance and efficiency of CPUs every year has created the need for turning computers into heterogeneous systems, i.e., systems composed of multiple types of processors that can suit better different types of workloads or parts of them. More than a decade ago, Graphics Processing Units (GPUs) became general-purpose parallel processors, in order to make their outstanding processing capabilities available to many workloads beyond graphics. GPUs have been a critical key to the recent rise of Machine Learning and Artificial Intelligence, which took unrealistic training times before the use of GPUs. Field-Programmable Gate Arrays (FPGAs) are another example computing device that can deliver impressive benefits in terms of performance and energy efficiency. More specific examples are (1) a plethora of specialized accelerators (e.g., Tensor Processing Units for neural networks), and (2) near-data processing architectures (i.e., placing compute capabilities near or inside memory/storage).
Despite the great advances in the adoption of heterogeneous systems in recent years, there are still many challenges to tackle, for example:
- Heterogeneous implementations (using GPUs, FPGAs, TPUs) of modern applications from important fields such as bioinformatics, machine learning, graph processing, medical imaging, personalized medicine, robotics, virtual reality, etc.
- Scheduling techniques for heterogeneous systems with different general-purpose processors and accelerators, e.g., kernel offloading, memory scheduling, etc.
- Workload characterization and programming tools that enable easier and more efficient use of heterogeneous systems.
If you are enthusiastic about working hands-on with different software, hardware, and architecture projects for heterogeneous systems, this is your P&S. You will have the opportunity to program heterogeneous systems with different types of devices (CPUs, GPUs, FPGAs, TPUs), propose algorithmic changes to important applications to better leverage the compute power of heterogeneous systems, understand different workloads and identify the most suitable device for their execution, design optimized scheduling techniques, etc. In general, the goal will be to reach the highest performance reported for a given important application.
Prerequisites of the course:
- Digital Design and Computer Architecture (or equivalent course).
- Familiarity with C/C++ programming and strong coding skills.
- Interest in future computer architectures and computing paradigms.
- Interest in discovering why things do or do not work and solving problems
- Interest in making systems efficient and usable
The course is conducted in English.
The course has two main parts:
1. Weekly lectures on GPU and heterogeneous programming.
2. Hands-on project: Each student develops his/her own project.
Mentors
Name | Office | ||
---|---|---|---|
Lead Supervisor | Juan Gómez Luna | juan.gomez@safari.ethz.ch | ETZ H 61.1 |
Supervisor | Mohammed Alser | alserm@ethz.ch | ETZ H 61.1 |
Supervisor | Behzad Salami | bsalami@ethz.ch | ETZ H 64 |
Supervisor | Mohammad Sadr | ETZ H 61.1 | |
Supervisor | Joel Lindegger | ETZ H 64 |
Lecture Video Playlist on YouTube
Spring 2023 Meetings/Schedule
Week | Date | Livestream | Meeting | Learning Materials | Assignments |
---|---|---|---|---|---|
W1 | 10.03 Fri. | Livestream | M1: P&S Course Presentation (PDF) (PPT) | Required Materials Recommended Materials | HW 0 Out |
W2 | 17.03 Fri. | Premiere | M2: SIMD Processing and GPUs (PDF) (PPT) | ||
Hands-on Project Proposals | |||||
W3 | 24.03 Fri. | Premiere | M3: GPU Software Hierarchy (PDF) (PPT) | ||
W4 | 31.03 Fri. | Premiere | M4: GPU Memory Hierarchy (PDF) (PPT) | ||
W5 | 07.04 Fri. | Premiere | M5: GPU Performance Considerations (PDF) (PPT) | ||
W6 | 14.04 Fri. | Premiere | M6: Parallel Patterns: Reduction (PDF) (PPT) | ||
W7 | 21.04 Fri. | Premiere | M7: Parallel Patterns: Histogram (PDF) (PPT) | ||
W8 | 28.04 Fri. | Premiere | M8: Parallel Patterns: Convolution (PDF) (PPT) | ||
W9 | 05.05 Fri. | Premiere | M9: Advanced Tiling for Matrix Multiplication (PDF) (PPT) | ||
W10 | 12.05 Fri. | Premiere | M10: Parallel Patterns: Prefix Sum (Scan) (PDF) (PPT) | ||
W11 | 19.05 Fri. | Premiere | M11: Parallel Patterns: Sparse Matrices (PDF) (PPT) | ||
W12 | 26.05 Fri. | Premiere | M12: Parallel Patterns: Graph Search (PDF) (PPT) | ||
W13 | 02.06 Fri. | Premiere | M13: Parallel Patterns: Merge (PDF) (PPT) | ||
W14 | 09.06 Fri. | Premiere | M14: Dynamic Parallelism (PDF) (PPT) | ||
W15 | 16.06 Fri. | Premiere | M15: Collaborative Computing (PDF) (PPT) |
Past Lecture Video Playlists on YouTube
Learning Materials
Meeting 1: Required Materials
- An introduction to SIMD processors and GPUs:
- An introduction to GPUs and heterogeneous programming:
Meeting 1: Recommended Materials
- Example recent studies of FPGA and GPU implementation for bioinformatics:
- An example recent study of a suite of heterogeneous benchmarks:
More Learning Materials
- An example recent study of a medical image application on GPU:
- Example studies of programming tools and performance portability on heterogeneous systems:
- Example studies of scheduling techniques for heterogeneous systems: