Please see Try an Example with IRIS
This is the prototype Python front end for the SPIRAL project's SnowWhite system, which compiles high-level specifications of numerical computations into hardware-specific optimized code. It supports a variety of CPU's as well as Nvidia and AMD GPU's on systems running Linux, Windows, or MacOS.
SnowWhite was developed under the DARPA PAPPA (Performant Automation of Parallel Program Assembly) program. The program focused on ways reduce the complexity of building software that takes advantage of the massive parallelism of advanced high-preformance computing systems.
SnowWhite, implemented as a Python module, uses the SPIRAL code generation system to translate NumPy-based specifications to generated code, then compiles that code into a loadable library.
See the examples directory to learn more.
DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.
Besides a current version of Python3 and NumPy, SnowWhite needs the following:
- SPIRAL (available on GitHub)
- spiral-software https://www.github.com/spiral-software/spiral-software
- spiral-package-fftx https://www.github.com/spiral-software/spiral-package-fftx
- spiral-package-simt https://www.github.com/spiral-software/spiral-package-simt
- CMake (3.14 or higher is required)
- C Compiler
With SPIRAL installed you will have CMake, a compatible C compiler, and Python3 for SnowWhite. SPIRAL builds on Linux/Unix with gcc and make, on Windows it builds with Visual Studio. For macOS SPIRAL requires version 10.14 (Mojave) or later of macOS, with a compatible version of Xcode and and Xcode Command Line Tools.
Clone spiral-software to a location on you computer. For example:
cd ~/work
git clone https://www.github.com/spiral-software/spiral-software
This location is known as SPIRAL HOME and you must set an environment variable SPIRAL_HOME to point to this location later.
To install the two spiral packages do the following:
cd ~/work/spiral-software/namespaces/packages
git clone https://www.github.com/spiral-software/spiral-package-fftx fftx
git clone https://www.github.com/spiral-software/spiral-package-simt simt
NOTE: The spiral packages must be installed under directory $SPIRAL_HOME/namespaces/packages and must be placed in folders with the prefix spiral-package removed.
Follow the build instructions for spiral-software.
Clone python-package-snowwhite to a location on you computer renaming it to snowwhite. For example:
cd ~/work
git clone https://github.com/spiral-software/python-package-snowwhite snowwhite
Add the directory that contains the snowwhite clone to the environment variable PYTHONPATH. (In the above example that would be ~/work.) This allows Python to locate the snowwhite module.
Copy one of the example Python scripts from the examples directory to a scratch directory and run it like this:
D:\Temp>python3 run-hockney8.py
...
...
Install the project...
-- Install configuration: "Release"
-- Installing: /path/to/spiral-software/snowwhite/.libs/libhockney_8_3_5.so
Max Diff between Python/C = 5.204170427930421e-18
The first time you run it, you will see output from the CMake/SPIRAL/C build, but after that it will run much faster using the generated library, which is placed in the snowwhite/.libs directory.
Some of the examples require additional arguments, and some options you can change. Read through the examples for better understanding, and examine the generated intermediate files in your scratch directory.
Install IRIS in your system. Please see https://github.com/ORNL/iris
Running an example on a NVIDIA CUDA system:
cd ~/work/snowwhite/example-iris
./1-setup-cuda.sh
./2-build-cuda.sh
./3-run-snowwhite.sh
./4-run-iris.sh
Running an example on an AMD HIP system:
cd ~/work/snowwhite/example-iris
./1-setup-hip.sh
./2-build-hip.sh
./3-run-snowwhite.sh
./4-run-iris.sh