|
| 1 | +{ |
| 2 | + "cells": [ |
| 3 | + { |
| 4 | + "cell_type": "markdown", |
| 5 | + "id": "2f93e5f6", |
| 6 | + "metadata": {}, |
| 7 | + "source": [ |
| 8 | + "## PyHPC - NumPy, CuPy, & mpi4py - 4 Hours\n", |
| 9 | + "\n", |
| 10 | + "### Notebooks\n", |
| 11 | + "\n", |
| 12 | + "| # | Topic | Solution | Technologies | \n", |
| 13 | + "|---|-------|----------|--------------|\n", |
| 14 | + "| 01 | [NumPy Intro - ndarray Basics](../fundamentals/01__numpy_intro__ndarray_basics.ipynb) | [Click](../fundamentals/solutions/01__numpy_intro__ndarray_basics__SOLUTION.ipynb) | NumPy |\n", |
| 15 | + "| 05 | [Memory Spaces - Power Iteration](../fundamentals/05__memory_spaces__power_iteration.ipynb) | [Click](../fundamentals/solutions/05__memory_spaces__power_iteration__SOLUTION.ipynb) | CuPy |\n", |
| 16 | + "| 06 | [Asynchrony - Power Iteration](../fundamentals/06__asynchrony__power_iteration.ipynb) | [Click](../fundamentals/solutions/06__asynchrony__power_iteration__SOLUTION.ipynb) | CuPy |\n", |
| 17 | + "| 60 | [mpi4py](../distributed/60__mpi4py.ipynb) | - | mpi4py |\n", |
| 18 | + "\n", |
| 19 | + "### Materials\n", |
| 20 | + "\n", |
| 21 | + "| Docker Compose | Brev Instance | Brev Provider |\n", |
| 22 | + "|----------------|---------------|---------------|\n", |
| 23 | + "| [Link](https://github.com/NVIDIA/accelerated-computing-hub/blob/generated/main/tutorials/accelerated-python/notebooks/syllabi/pyhpc__numpy_cupy_mpi4py__4_hours__docker_compose.yml) | 4xL4, 2xL4, 2xL40S, or 1x L40S | Crusoe or any other with Flexible Ports |\n", |
| 24 | + "\n", |
| 25 | + "### Introduction\n", |
| 26 | + "\n", |
| 27 | + "In this tutorial we will cover:\n", |
| 28 | + "- NumPy fundamentals and operations for scientific computing.\n", |
| 29 | + "- Accelerating NumPy workflows with CuPy on GPUs\n", |
| 30 | + "- Understanding memory spaces and data movement between CPU and GPU\n", |
| 31 | + "- Leveraging asynchronous operations for better performance\n", |
| 32 | + "- Distributed computing with mpi4py for multi-node parallelism\n", |
| 33 | + "\n", |
| 34 | + "Attendees will be expected to have a general knowledge of HPC and Python programming concepts. Some familiarity with NumPy is helpful but not required.\n" |
| 35 | + ] |
| 36 | + } |
| 37 | + ], |
| 38 | + "metadata": { |
| 39 | + "colab": { |
| 40 | + "gpuType": "T4", |
| 41 | + "provenance": [] |
| 42 | + }, |
| 43 | + "kernelspec": { |
| 44 | + "display_name": "Python 3 (ipykernel)", |
| 45 | + "language": "python", |
| 46 | + "name": "python3" |
| 47 | + }, |
| 48 | + "language_info": { |
| 49 | + "name": "python" |
| 50 | + } |
| 51 | + }, |
| 52 | + "nbformat": 4, |
| 53 | + "nbformat_minor": 5 |
| 54 | +} |
0 commit comments