Reorganise structure

EPCCed · May 2, 2023 · 6d7ca21 · 6d7ca21
1 parent 7788b2d
commit 6d7ca21
Show file tree

Hide file tree

Showing 334 changed files with 155 additions and 155 deletions.
diff --git a/README.md b/README.md
@@ -1,28 +1,139 @@
-# ARCHER Introduction to GPU Programming
 
-This repository contains both the slides and the exercise material
-for ARCHER and Cirrus Introdution to GPU Programming Course.
+<img src="./images/archer2_logo.png" align="left" width="355" height="100" />
+<img src="./images/epcc_logo.jpg" align="right" width="133" height="100" />
 
-ARCHER is the United Kingdom National Supercomputing service
-http://www.archer.ac.uk/
+<br><br><br><br>
 
-Cirrus is a United Kingdom Tier2 service https://www.cirrus.ac.uk
+# Introduction to GPU programming with CUDA/HIP
 
-The latest slides for the lecture content are available at:
+[![CC BY-NC-SA 4.0][cc-by-nc-sa-shield]][cc-by-nc-sa]
 
-https://epcced.github.io/archer-gpu-course/
+This short course will provide an introduction to GPU computing with CUDA
+aimed at scientific application programmers wishing to develop their own
+software. The course will give a background on the difference between CPU
+and GPU architectures as a prelude to introductory exercises in CUDA
+programming. The course will discuss the execution of kernels, memory
+management, and shared memory operations. Common performance issues are
+discussed and their solution addressed. Profiling will be introduced via
+the current NVIDIA tools.
 
-The slides are built using [reveal.js](https://github.com/hakimel/reveal.js).
+The course will go on to consider execution of independent streams, and
+the execution of work composed as a collection of dependent tasks expressed
+as a graph. Device management and details of device to device data transfer
+will be covered for situations where more than one GPU device is available.
+CUDA-aware MPI will be covered.
 
-If you would like to save a pdf copy of the slides, please follow the
-instructions at https://github.com/hakimel/reveal.js/#pdf-export
+The course will not discuss programming with compiler directives, but does
+provide a concrete basis of understanding of the underlying principles of
+the CUDA model which is useful for programmers ultimately wishing to make
+use of OpenMP or OpenACC (or indeed other models). The course will not
+consider graphics programming, nor will it consider machine learning
+packages.
 
+Note that the course is also appropriate for those wishing to use AMD GPUs
+via the HIP API, although we will not specifically use HIP.
 
-#### License
+Attendees must be able to program in C or C++ (course examples and
+exercises will limit themselves to C). A familiarity with threaded
+programming models would be useful, but no previous knowledge of GPU
+programming is required.
 
-Slides are available under a Creative Commons license.
+## Installation
 
-Other material is Copyright (c) EPCC, The University of Edinburgh, unless
-otherwise stated.
+For details of how to log into a Cirrus account, see
+https://cirrus.readthedocs.io/en/main/user-guide/connecting.html
 
-Kokkos exercises are based on Tutorial material by Sandia and are (c) Sandia.
+Check out the git repository to your Cirrus account.
+```
+$ cd ${HOME/home/work}
+$ https://github.com/EPCCed/archer-gpu-course.git
+$ cd archer-gpu-course
+```
+For the examples and exercises in the course, we will use the
+NVIDIA compiler driver `nvcc`. To access this
+```
+$ module load nvidia/nvhpc
+```
+Check you can compile and run a very simple program
+and submit the associated script to the queue system.
+```
+$ cd section-2.01
+$ nvcc -arch=sm_70 exercise_dscal.cu
+$ sbatch submit.sh
+```
+The result should appear in a file `slurm-123456.out` in the working
+directory.
+
+Each section of the course is associated with a different directory, each
+of which contains a number of example programs and exercise templates.
+Answers to exercises generally re-appear as templates to later exercises.
+Miscellaneous solutions also appear in the solutions directory.
+
+
+## Timetable
+
+The timetable may shift slightly in terms of content, but we will stick to
+the advertised start and finish times, and the break times.
+
+
+### Day one
+
+| Time  | Content                                  | Section                      |
+|-------|------------------------------------------|------------------------------|
+| 09:30 | Logistics, login, modules, local details | See above                    |
+| 10:00 | Introduction                             |                              |
+|       | Performance model; Graphics processors   | [section-1.01](section-1.01) |
+| 10:30 | The CUDA/HIP programming model           |                              |
+|       | Abstraction; host code and device code   | [section-1.02](section-1.02) |
+| 11:00 | Break                                    |                              |
+| 11:30 | CUDA/HIP programming                     |                              |
+|       | Memory management, exercise              | [section-2.01](section-2.01) |
+| 12:15 | CUDA/HIP programming (cont.)             |                              |
+|       | Kernels, exercise                        | [section-2.02](section-2.02) |
+| 13:00 | Lunch                                    |                              |
+| 14:00 | Some performance considerations          |                              |
+|       | Exercise on matrix operation             | [section-2.03](section-2.03) |
+| 15:00 | Break                                    |                              |
+| 15:20 | More on memory: managed memory           |                              |
+|       | Exercise on managed memory               | [section-2.04](section-2.04) |
+| 15:50 | More on memory: shared memory            |                              |
+| 16:10 | Exercise on vector product               | [section-2.05](section-2.05) |
+| 16:30 | All together: matrix-vector product      | [][]                         |
+| 17:00 | Close                                    |                              |
+
+
+### Day two
+
+
+| Time  | Content                                  | Section                      |
+|-------|------------------------------------------|------------------------------|
+| 09:00 | Detour: visual profiler                  |                              |
+| 09:10 | Exercise: nsight systems and compute     | [section-3.01](section-3.01)      |
+| 09:30 | Streams                                  |                              |
+|       | Using `cudaMempcyAsync()` etc            | [section-4.01](section-4.01) |
+| 10:00 | Graph API                                |                              |
+|       | Using `cudaGraphLaunch()` etc            | [section-4.02](section-4.02) |
+| 11:00 | Break                                    |                              |
+| 11:30 | Device management: more then one GPU     |                              |
+|       | `cudaMemcpy()` again                     | [section-5.01](section-5.01) |
+| 12:15 | Special topic: GPU-aware MPI             |                              |
+|       | Exercise                                 | [section-5.02](section-5.02) |
+| 13:00 | Lunch                                    |                              |
+| 14:00 | Putting it all together                  |                              |
+|       | Conjugate gradient exercise              | [section-6.01](section-6.01) |
+| 15:00 | Break                                    |                              |
+| 15:20 | Exercises                                |                              |
+| 15:50 | Miscellaneous comments                   | [section-7.01](section-7.01) |
+| 16:00 | Close                                    |                              |
+
+
+
+---
+This work is licensed under a
+[Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][cc-by-nc-sa].
+
+[cc-by-nc-sa]: http://creativecommons.org/licenses/by-nc-sa/4.0/
+[cc-by-nc-sa-image]: https://licensebuttons.net/l/by-nc-sa/4.0/88x31.png
+[cc-by-nc-sa-shield]: https://img.shields.io/badge/License-CC%20BY--NC--SA%204.0-lightgrey.svg
+
+[![CC BY-NC-SA 4.0][cc-by-nc-sa-image]][cc-by-nc-sa]
diff --git a/applications-two-day/README.md b/applications-two-day/README.md
diff --git a/applications-two-day/images/archer2_logo.png → images/archer2_logo.png b/applications-two-day/images/archer2_logo.png → images/archer2_logo.png
diff --git a/applications-two-day/images/epcc_logo.jpg → images/epcc_logo.jpg b/applications-two-day/images/epcc_logo.jpg → images/epcc_logo.jpg
diff --git a/...o-day/images/ks-schematic-host-device.svg → images/ks-schematic-host-device.svg b/...o-day/images/ks-schematic-host-device.svg → images/ks-schematic-host-device.svg
diff --git a/...y/images/ks-schematic-memory-transfer.svg → images/ks-schematic-memory-transfer.svg b/...y/images/ks-schematic-memory-transfer.svg → images/ks-schematic-memory-transfer.svg
diff --git a/...ns-two-day/images/ks-schematic-simple.svg → images/ks-schematic-simple.svg b/...ns-two-day/images/ks-schematic-simple.svg → images/ks-schematic-simple.svg
diff --git a/...o-day/images/ks-threads-blocks-grids.jpeg → images/ks-threads-blocks-grids.jpeg b/...o-day/images/ks-threads-blocks-grids.jpeg → images/ks-threads-blocks-grids.jpeg
diff --git a/...ons-two-day/images/ks-threads-blocks.jpeg → images/ks-threads-blocks.jpeg b/...ons-two-day/images/ks-threads-blocks.jpeg → images/ks-threads-blocks.jpeg
diff --git a/applications-two-day/images/ks-threads.jpeg → images/ks-threads.jpeg b/applications-two-day/images/ks-threads.jpeg → images/ks-threads.jpeg
diff --git a/one-day/README.md b/one-day/README.md
@@ -0,0 +1,28 @@
+# ARCHER Introduction to GPU Programming
+
+This repository contains both the slides and the exercise material
+for ARCHER and Cirrus Introdution to GPU Programming Course.
+
+ARCHER is the United Kingdom National Supercomputing service
+http://www.archer.ac.uk/
+
+Cirrus is a United Kingdom Tier2 service https://www.cirrus.ac.uk
+
+The latest slides for the lecture content are available at:
+
+https://epcced.github.io/archer-gpu-course/
+
+The slides are built using [reveal.js](https://github.com/hakimel/reveal.js).
+
+If you would like to save a pdf copy of the slides, please follow the
+instructions at https://github.com/hakimel/reveal.js/#pdf-export
+
+
+#### License
+
+Slides are available under a Creative Commons license.
+
+Other material is Copyright (c) EPCC, The University of Edinburgh, unless
+otherwise stated.
+
+Kokkos exercises are based on Tutorial material by Sandia and are (c) Sandia.