julia-agu-2022
所属分类:GPU/显卡
开发工具:Jupyter Notebook
文件大小:1409KB
下载次数:0
上传日期:2022-12-10 15:20:21
上 传 者:
sh-1993
说明: 使用Julia和GPU编程进行数值建模
(Using Julia and GPU programming for numerical modelling)
文件列表:
LICENSE (1070, 2022-12-10)
Project.toml (358, 2022-12-10)
deploy_notebooks.jl (232, 2022-12-10)
figures (0, 2022-12-10)
figures\Stokes2D_vep.gif (583410, 2022-12-10)
figures\cpu_gpu_evo.png (102490, 2022-12-10)
figures\gpu.png (118222, 2022-12-10)
figures\ice2.png (239454, 2022-12-10)
figures\logo2.png (4351, 2022-12-10)
figures\mem_wall.png (164397, 2022-12-10)
figures\parallelstencil.png (24113, 2022-12-10)
figures\ps_igg_perf.png (73055, 2022-12-10)
figures\two_lang.png (136696, 2022-12-10)
julia-workshop-agu22.ipynb (28928, 2022-12-10)
julia-workshop-agu22.jl (21694, 2022-12-10)
# Julia AGU 2022 workshop
Using Julia and GPU programming for numerical modelling
“ **SCIWS26: Porting Machine Learning and Modeling from a Laptop to a Supercomputer**
## Automatic notebook generation
The presentation slides and the demo notebook are self-contained in a Jupyter notebook [julia-workshop-agu22.ipynb](https://github.com/luraess/julia-agu-2022/blob/master/julia-workshop-agu22.ipynb) that can be auto-generated using literate programming by deploying the [julia-workshop-agu22.jl](https://github.com/luraess/julia-agu-2022/blob/master/julia-workshop-agu22.jl) script.
To reproduce:
1. Clone this git repo
2. Open Julia and resolve/instantiate the project
```julia-repl
using Pkg
Pkg.activate(@__DIR__)
Pkg.resolve()
Pkg.instantiate()
```
3. Run the deploy script
```julia-repl
julia> using Literate
julia> include("deploy_notebooks.jl")
```
4. Then using IJulia, you can launch the notebook and get it displayed in your web browser:
```julia-repl
julia> using IJulia
julia> notebook(dir=pwd())
```
_To view the notebook as slide, you need to install the [RISE](https://github.com/luraess/julia-agu-2022/blob/master/https://rise.readthedocs.io/en/stable/installation.html) plugin_
## Resources
#### Packages
- [ParallelStencil.jl](https://github.com/luraess/julia-agu-2022/blob/master/https://github.com/omlins/ParallelStencil.jl)
- [ImplicitGlobalGrid.jl](https://github.com/luraess/julia-agu-2022/blob/master/https://github.com/eth-cscs/ImplicitGlobalGrid.jl)
#### Papers
- Bridging HPC Communities through the Julia Programming Language: https://doi.org/10.48550/ARXIV.2211.02740
- High-performance xPU Stencil Computations in Julia: https://doi.org/10.48550/arXiv.2211.15634
- Distributed Parallelization of xPU Stencil Computations in Julia: https://doi.org/10.48550/arXiv.2211.15716
#### Courses and resources
- ETHZ course on solving PDEs with GPUs: https://pde-on-gpu.vaw.ethz.ch
- Advanced GPU HPC optimisation course at CSCS: https://github.com/omlins/julia-gpu-course and https://github.com/maleadt/cscs_gpu_course/
- More [here](https://github.com/luraess/julia-agu-2022/blob/master/https://pde-on-gpu.vaw.ethz.ch/extras/#extra_material)
#### Misc
- Frontier GPU multi-physics solvers: https://ptsolvers.github.io/GPU4GEO/
近期下载者:
相关文件:
收藏者: