Title

DPC++ Reductions: This is part 8 of the oneAPI essentials training series

Requirements

Optimized for	Description
OS	Linux* Ubuntu 18.04, 20 Windows* 10
Hardware	Skylake with GEN9 or newer
Software	Intel® oneAPI DPC++ Compiler, Jupyter Notebooks, Intel Devcloud

Purpose

This hands-on exercise demonstrates various ways to optimizes reduction operations using DPC++. You will learn how to use ND-Range kernels to parallelize reductions on accelerators and also learn how to optimize reductions using new reduction extensions in DPC++.

License

Code samples are licensed under the MIT license. See License.txt for details.

Third party program Licenses can be found here: third-party-programs.txt

Install Directions

The Jupyter notebooks are tested and can be run on Intel Devcloud. Below are the steps to access these Jupyter notebooks on Intel Devcloud

1. Register on Intel Devcloud
2. Go to the "Terminal" in the Intel Devcloud
3. Type in the below command to download the oneAPI-essentials series notebooks into your Devcloud account /data/oneapi_workshop/get_jupyter_notebooks.sh
4. Navigate to oneAPI_Essentials folder and open the Welcome.ipynb, click on "DPC++ Reductions" notebook and follow the instructions