Comments (3)
Sorry for the late reply. CellArrays can be used for additional dimensions. Currently you need to use the main branch if you want to use this functionality. You can find very basic example here: https://omlins.github.io/CellArrays.jl/dev/examples/memcopyCellArray3D_ParallelStencil/
from parallelstencil.jl.
Thanks for the hint! I am not sure if I understood it correctly. Do I understand the example correctly that it shows a three dimensional array whose values can be represented in a 4x4 way? If so, I was asking more towards if its possible to get another parallel index and another dimension of the underlying array so something like
T = @zeros(nx, ny, nz, nu);
T2 = @zeros(nx, ny, nz, nu);
Ci = @zeros(nx, ny, nz, nu);
and
@parallel_indices (ix,iy,iz,iu) function copy3D_explicit!(T2::CellArray, T::CellArray, Ci::CellArray)
T2[ix,iy,iz,iu] = T[ix,iy,iz,iu] + Ci[ix,iy,iz,iu];
return
end
from parallelstencil.jl.
I'm not sure the example was fully clear to you. In the example it says:
nx, ny, nz = 128, 128, 1024
celldims = (4, 4)
T = @zeros(nx, ny, nz, celldims=celldims)
This will create a CellArray, which contains 128x128x1024 cells, where each cell consists of a matrix of 4x4 values. So, in the end it is a five dimensional array.
Operations on these cells follow the standard rules and syntax of array operations in julia.
if nevertheless you really need a 4th parallel index, then you can compute it yourself based on the one-dimensional, two-dimensional or three-dimensional indices. Just like in CUDA, only one-dimensional, two-dimensional or three-dimensional indices can be obtained directly. See for example: https://stackoverflow.com/questions/44368525/using-linear-index-to-map-to-4d-array
from parallelstencil.jl.
Related Issues (20)
- AMDGPU v0.5.0 compat HOT 1
- Add device_sync
- sync issues on AMDGPU backend
- Make CellArrays mutable HOT 4
- finite volume method HOT 3
- [JuliaCon/proceedings-review] @parallel keyword argument `loopopt` deprecated? HOT 1
- ParallelStencil on 1.10 HOT 6
- [JuliaCon/proceedings-review] DOI of paper by Besard et al. HOT 2
- [JuliaCon/proceedings-review] Community guidelines HOT 1
- [JuliaCon/proceedings-review] Performance metrics HOT 4
- Type unstable Data.Number HOT 2
- GPU memory management issue when running multi-GPU code HOT 10
- Add support for Polyester's `@batch` HOT 20
- Generalize loopopt
- Create and update GPU unit tests
- Thread (CPU) Float32/Float64 performance comparison on miniapp acoustic2D HOT 12
- Example for init_global_grid_usage HOT 3
- How to implement custom finite differencing operators HOT 8
- CUDA Crash with julia 1.9.0 HOT 8
- Non cartesian gather! HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from parallelstencil.jl.