• Plumed
• WESTPA

Tools and plugins to enhance molecular dynamics sampling

• Roadmaps in Cornell Virtual Workshop
• Search for topics

Cornell Virtual Workshop is a comprehensive training resource for high performance computing topics. The Cornell University Center for Advanced Computing (CAC) is a leader in the development and deployment of Web-based training programs. Our Cornell Virtual Workshop learning platform is designed to enhance the computational science skills of researchers, accelerate the adoption of new and emerging technologies, and broaden the participation of underrepresented groups in science and engineering. Over 350,000 unique visitors have accessed Cornell Virtual Workshop training on programming languages, parallel computing, code improvement, and data analysis. The platform supports learning communities around the world, with code examples from national systems such as Frontera, Stampede2, and Jetstream2.

• PyFR documentation
• PyFR source code from Github
• Discourse channel for discussions and help

What is PyFR and how does it solve fluid flow problems? PyFR is an open-source Computational Fluid Dynamics (CFD) solver that is based on Python and employs the high-order Flux Reconstruction technique. It effectively solves fluid flow problems by utilizing streaming architectures, making it suitable for complex fluid dynamics simulations. How does PyFR achieve scalability on clusters with CPUs and GPUs? PyFR achieves scalability by leveraging distributed memory parallelism through the Message Passing Interface (MPI). It implements persistent, non-blocking MPI requests using point-to-point (P2P) communication and organizes kernel calls to enable local computations while exchanging ghost states. This design approach allows PyFR to efficiently operate on clusters with heterogeneous architectures, combining CPUs and GPUs. Why is PyFR valuable for benchmarking clusters? PyFR's exceptional performance has been recognized by its selection as a finalist in the ACM Gordon Bell Prize for High-Performance Computing. It demonstrates strong-scaling capabilities by effectively utilizing low-latency inter-GPU communication and achieving strong-scaling on unstructured grids. PyFR has been successfully benchmarked with up to 18,000 NVIDIA K20X GPUs on Titan, showcasing its efficiency in handling large-scale simulations.

• GPU Acceleration in Python

This tutorial explains how to use Python for GPU acceleration with libraries like CuPy, PyOpenCL, and PyCUDA. It shows how these libraries can speed up tasks like array operations and matrix multiplication by using the GPU. Examples include replacing NumPy with CuPy for large datasets and using PyOpenCL or PyCUDA for more control with custom GPU kernels. It focuses on practical steps to integrate GPU acceleration into Python programs.

• Introduction to Parallel Programming for GPUs with CUDA

This tutorial provides a comprehensive introduction to CUDA programming, focusing on essential concepts such as CUDA thread hierarchy, data parallel programming, host-device heterogeneous programming model, CUDA kernel syntax, GPU memory hierarchy, and memory optimization techniques like global memory coalescing and shared memory bank conflicts. Aimed at researchers, students, and practitioners, the tutorial equips participants with the skills needed to leverage GPU acceleration for scalable computation, particularly in the context of AI.

• PyFR installation to local machine

These instructions were executed on the FASTER and Grace cluster computing facilities at Texas A&M University. However, the process can be applied to other clusters with similar environments. For local installation, please refer to the PyFR documentation. Please note that these instructions were valid at the time of writing. Depending on the time you're executing these, the versions of the modules may need to be updated. 1. Loading Modules The first step involves loading pre-installed software libraries required for PyFR. Execute the following commands in your terminal to load these modules: module load foss/2022b module load libffi/3.4.4 module load OpenSSL/1.1.1k module load METIS/5.1.0 module load HDF5/1.13.1 2. Python Installation from Source Choose a location for Python 3.11.1 installation, preferably in a .local directory. Navigate to the directory containing the Python 3.11.1 source code. Then configure and install Python: cd $INSTALL/Python-3.11.1/ ./configure --prefix=$LOCAL --enable-shared --with-system-ffi --with-openssl=/sw/eb/sw/OpenSSL/1.1.1k-GCCcore-11.2.0/ PKG_CONFIG_PATH=$LOCAL/pkgconfig LDFLAGS=/usr/lib64/libffi.so.6.0.2 make clean; make -j20; make install; 3. Virtual Environment Setup A virtual environment allows you to isolate Python packages for this project from others on your system. Create and activate a virtual environment using: pip3.11 install virtualenv python3.11 -m venv pyfr-venv . pyfr-venv/bin/activate 4. Install PyFR Dependencies Several Python packages are required for PyFR. Install these packages using the following commands: pip3 install --upgrade pip pip3 install --no-cache-dir wheel pip3 install --no-cache-dir botorch pandas matplotlib pyfr pip3 uninstall -y pyfr 5. Install PyFR from Source Finally, navigate to the directory containing the PyFR source code, and then install PyFR: cd /scratch/user/sambit98/github/PyFR/ python3 setup.py develop Congratulations! You've successfully set up PyFR on the FASTER and Grace cluster computing facilities. You should now be able to use PyFR for your computational fluid dynamics simulations.

• thrust_ex.txt

Some examples for writing Thrust code. To compile, download the CUDA compiler from NVIDIA. This code was tested with CUDA 9.2 but is likely compatible with other versions. Before compiling change extension from thrust_ex.txt to thrust_ex.cu. Any code on the device (GPU) that is run through a Thrust transform is automatically parallelized on the GPU. Host (CPU) code will not be. Thrust code can also be compiled to run on a CPU for practice.