xfygogo
/
phasicFlow
mirror of https://github.com/PhasicFlow/phasicFlow.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'PhasicFlow:main' into main

This commit is contained in:
Simboy 2025-03-06 18:39:55 +08:00 committed by GitHub
parent eb62adc87d 169dd73963
commit 84197bf237
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 52 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

86
README.md
View File

@ -1,48 +1,66 @@
<div align ="center">
<img src="doc/phasicFlow_logo_github.png" style="width: 400px;">
<div align="center">
<img src="doc/phasicFlow_logo_github.png" style="width: 400px;" alt="PhasicFlow Logo">
</div>
## **PhasicFlow: High-Performance Discrete Element Method Simulations**
**PhasicFlow** is a parallel C++ code for performing DEM simulations. It can run on shared-memory multi-core computational units such as multi-core CPUs or GPUs (for now it works on CUDA-enabled GPUs). The parallelization method mainly relies on loop-level parallelization on a shared-memory computational unit. You can build and run PhasicFlow in serial mode on regular PCs, in parallel mode for multi-core CPUs, or build it for a GPU device to off-load computations to a GPU. In its current statues you can simulate millions of particles (up to 80M particles tested) on a single desktop computer. You can see the [performance tests of PhasicFlow](https://github.com/PhasicFlow/phasicFlow/wiki/Performance-of-phasicFlow) in the wiki page.
PhasicFlow is a robust, open-source C++ framework designed for the efficient simulation of granular materials using the Discrete Element Method (DEM). Leveraging parallel computing paradigms, PhasicFlow is capable of executing simulations on shared-memory multi-core architectures, including CPUs and NVIDIA GPUs (CUDA-enabled). The core parallelization strategy focuses on loop-level parallelism, enabling significant performance gains on modern hardware. Users can seamlessly transition between serial execution on standard PCs, parallel execution on multi-core CPUs (OpenMP), and accelerated simulations on GPUs. Currently, PhasicFlow supports simulations involving up to 80 million particles on a single desktop workstation. Detailed performance benchmarks are available on the [PhasicFlow Wiki](https://github.com/PhasicFlow/phasicFlow/wiki/Performance-of-phasicFlow).
**MPI** parallelization with dynamic load balancing is under development. With this level of parallelization, PhasicFlow can leverage the computational power of **multi-gpu** workstations or clusters with distributed memory CPUs.
In summary PhasicFlow can have 6 execution modes:
1. Serial on a single CPU core,
2. Parallel on a multi-core computer/node (using OpenMP),
3. Parallel on an nvidia-GPU (using Cuda),
4. Parallel on distributed memory workstation (Using MPI)
5. Parallel on distributed memory workstations with multi-core nodes (using MPI+OpenMP)
6. Parallel on workstations with multiple GPUs (using MPI+Cuda).
## How to build?
You can build PhasicFlow for CPU and GPU executions. The latest release of PhasicFlow is v-0.1. [Here is a complete step-by-step procedure for building phasicFlow-v-0.1.](https://github.com/PhasicFlow/phasicFlow/wiki/How-to-Build-PhasicFlow).
**Scalable Parallelism: MPI Integration**
## Online code documentation
You can find a full documentation of the code, its features, and other related materials on [online documentation of the code](https://phasicflow.github.io/phasicFlow/)
Ongoing development includes the integration of MPI-based parallelization with dynamic load balancing. This enhancement will extend PhasicFlow's capabilities to distributed memory environments, such as multi-GPU workstations and high-performance computing clusters. Upon completion, PhasicFlow will offer six distinct execution modes:
## How to use PhasicFlow?
You can navigate into [tutorials folder](./tutorials) in the phasicFlow folder to see some simulation case setups. If you need more detailed discription, visit our [wiki page tutorials](https://github.com/PhasicFlow/phasicFlow/wiki/Tutorials).
1. **Serial Execution:** Single-core CPU.
2. **Shared-Memory Parallelism:** Multi-core CPU (OpenMP).
3. **GPU Acceleration:** NVIDIA GPU (CUDA).
4. **Distributed-Memory Parallelism:** MPI.
5. **Hybrid Parallelism:** MPI + OpenMP.
6. **Multi-GPU Parallelism:** MPI + CUDA.
## [PhasicFlowPlus](https://github.com/PhasicFlow/PhasicFlowPlus)
PhasicFlowPlus is and extension to PhasicFlow for simulating particle-fluid systems using resolved and unresolved CFD-DEM. [See the repository of this package.](https://github.com/PhasicFlow/PhasicFlowPlus)
# **Build and Installation**
PhasicFlow can be compiled for both CPU and GPU execution.
* **Current Development (v-1.0):** Comprehensive build instructions are available [here](https://github.com/PhasicFlow/phasicFlow/wiki/How-to-build-PhasicFlow%E2%80%90v%E2%80%901.0).
* **Latest Release (v-0.1):** Detailed build instructions are available [here](https://github.com/PhasicFlow/phasicFlow/wiki/How-to-Build-PhasicFlow).
# **Comprehensive Documentation**
In-depth documentation, including code structure, features, and usage guidelines, is accessible via the [online documentation portal](https://phasicflow.github.io/phasicFlow/).
## **Tutorials and Examples**
Practical examples and simulation setups are provided in the [tutorials directory](./tutorials). For detailed explanations and step-by-step guides, please refer to the [tutorial section on the PhasicFlow Wiki](https://github.com/PhasicFlow/phasicFlow/wiki/Tutorials).
# **PhasicFlowPlus: Coupled CFD-DEM Simulations**
PhasicFlowPlus is an extension of PhasicFlow that facilitates the simulation of particle-fluid systems using resolved and unresolved CFD-DEM methods. The repository for PhasicFlowPlus can be found [here](https://github.com/PhasicFlow/PhasicFlowPlus).
## Supporting packages
* [Kokkos](https://github.com/kokkos/kokkos) from National Technology & Engineering Solutions of Sandia, LLC (NTESS)
* [CLI11 1.8](https://github.com/CLIUtils/CLI11) from University of Cincinnati.
# How to cite PhasicFlow?
## How to cite PhasicFlow
If you are using PhasicFlow in your research or industrial work, cite the following [article](https://www.sciencedirect.com/science/article/pii/S0010465523001662):
```
@article{NOROUZI2023108821,
title = {PhasicFlow: A parallel, multi-architecture open-source code for DEM simulations},
journal = {Computer Physics Communications},
volume = {291},
pages = {108821},
year = {2023},
issn = {0010-4655},
doi = {https://doi.org/10.1016/j.cpc.2023.108821},
url = {https://www.sciencedirect.com/science/article/pii/S0010465523001662},
author = {H.R. Norouzi},
keywords = {Discrete element method, Parallel computing, CUDA, GPU, OpenMP, Granular flow}
@article
{
NOROUZI2023108821,
title = {PhasicFlow: A parallel, multi-architecture open-source code for DEM simulations},
journal = {Computer Physics Communications},
volume = {291},
pages = {108821},
year = {2023},
issn = {0010-4655},
doi = {https://doi.org/10.1016/j.cpc.2023.108821},
url = {https://www.sciencedirect.com/science/article/pii/S0010465523001662},
author = {H.R. Norouzi},
keywords = {Discrete element method, Parallel computing, CUDA, GPU, OpenMP, Granular flow}
}
```
# **Dependencies**
PhasicFlow relies on the following external libraries:
* **Kokkos:** A performance portability ecosystem developed by National Technology & Engineering Solutions of Sandia, LLC (NTESS). ([https://github.com/kokkos/kokkos](https://github.com/kokkos/kokkos))
* **CLI11 1.8:** A command-line interface parser developed by the University of Cincinnati. ([https://github.com/CLIUtils/CLI11](https://github.com/CLIUtils/CLI11))