cuda and ka

Author: Rabab Alomairy

parts/gpu/README.md

@@ -0,0 +1,64 @@
+# GPU Acceleration
+
+This section will teach you GPU programming in Julia, with a focus on using CUDA.jl and KernelAbstractions.jl.
+
+To run the notebooks on Perlmutter, please clone this repository if you haven't done so already:
+
+```sh
+git clone https://github.com/JuliaParallel/julia-hpc-tutorial-sc24
+cd julia-hpc-tutorial-sc24
+```
+
+It is recommended to access the Jupyter notebook for this section through this website:   
+
+[NERSC Jupyter notebook](https://jupyter.nersc.gov/)
+
+## Introduction to GPU Programming in Julia
+
+If you're looking for an introductory overview of GPU programming in Julia or GPU programming concepts in general, feel free to explore this [notebook](https://github.com/JuliaParallel/julia-hpc-tutorial-sc24/blob/main/parts/gpu/gpu_introduction.ipynb).
+
+## Heat Diffusion Simulation Using Julia
+
+This [notebook](https://github.com/JuliaParallel/julia-hpc-tutorial-sc24/blob/main/parts/gpu/heat-diffusion.ipynb) demonstrates the implementation of a 2D heat diffusion model using Julia, showcasing GPU acceleration with CUDA.jl and support for both CPU and GPU execution using KernelAbstractions.jl. It highlights efficient solutions for partial differential equations (PDEs).
+
+### Benchmarking Results
+
+The notebook presents a series of benchmarks comparing execution times across different configurations:
+
+1. **CPU Implementation**
+   - Utilizes Julia's native array operations.
+   - Serves as a baseline for performance comparison.
+
+2. **GPU Implementation with `CUDA.jl`**
+   - Employs `CUDA.jl` for direct GPU programming.
+   - Demonstrates significant speedup over the CPU version.
+
+3. **GPU Implementation with `KernelAbstractions.jl`**
+   - Uses `KernelAbstractions.jl` to write code that can run on both CPU and GPU.
+   - Offers flexibility with performance close to the `CUDA.jl` implementation.
+
+Here's the plot showing the benchmark results on a single A100 GPU in Perlmutter using problem of size `10240x10240` and number of time steps `2000`:
+
+
+ <img src="img/benchmarking.png" alt="Benchmark Results" width="500">
+
+
+### Key Takeaways
+
+- **Performance Gains:** GPU implementations, both with `CUDA.jl` and `KernelAbstractions.jl`, exhibit substantial performance improvements over the CPU version, highlighting the advantages of GPU acceleration for computationally intensive tasks like heat diffusion.
+
+- **Flexibility vs. Performance:** While `CUDA.jl` provides optimal performance for NVIDIA GPUs, `KernelAbstractions.jl` offers a more flexible approach, allowing code to run on multiple backends (CPU, GPU) with minimal changes, albeit with a slight performance trade-off.
+
+- **Ease of Use:** Both packages integrate seamlessly with Julia, enabling efficient development of high-performance applications without sacrificing code readability or maintainability.
+
+
+##  Gray-Scott Reaction-Diffusion Model Using Julia
+This [notebook](https://github.com/JuliaParallel/julia-hpc-tutorial-sc24/blob/main/parts/gpu/gray-scott.ipynb) introduces Gray-Scott Reaction-Diffusion model using Julia. 
+ 
+### Example Plots
+
+The notebook showcases the evolution of the Gray-Scott model over time, illustrating the intricate patterns generated by the reaction-diffusion process. Here are some example plots:
+
+<img src="img/gs-plain.png" alt="Benchmark Results" width="600">
+<img src="img/gs-small.png" alt="Benchmark Results" width="600">
+<img src="img/gs-funky.png" alt="Benchmark Results" width="600">