GPUSPH

GPUSPH v5 released

Thu, 13 Jun 2019 16:00:00 +0200

GPUSPH version 5.0 is out! Head over to the release tag on GitHub.

Dear all,

we are pleased to announce the official release of GPUSPH version 5.0. The new release collects two years of development including new physical and numerical models, several bug fixes, better performance, and higher flexibility for the implementation of new features and integration schemes.

Major changes of note include:

possibility to create GPUSPH simulation scenarios using the SALOME platform;
Hu & Adams formulation for multi-fluid support;
additional models for the viscous contribution to the Navier–Stokes equations (Monaghan, Español & Revenga);
more sophisticated specification of the viscous model options, including separate specifications for the turbulence model, averaging operator and fluid rheology;
support for several non-Newtonian rheologies including Bingham plastic, granular rheology, and the Herschel–Bulkley and DeKee & Turcotte rheological models;
versioned interface to define simulation scenarios, with XProblem corresponding to Problem API version 1;
experimental support for a repacking phase before the simulation starts.

Requirements:

operating systems:
- Linux is officially supported and tested;
- Mac OS X support has not been tested with this release; support for this operating system may be phased out in the future, due to the difficulty in keeping up with the evolution of the CUDA and compiler support on that platform;
- experimental support for Windows is in the works, but it has not been integrated into version 5 due to known bugs in the Microsoft compiler;
CUDA 7.0 or later is needed; CUDA 8.0 or later is recommended;
a host compiler with C++11 support, and supported by CUDA, is needed:
- on Linux, GCC 4.8 or later, or Clang 3.6 or later (CUDA 7.5 or
- later only); on Mac OS X, XCode 6 or later;
Project Chrono, for fully coupled fluid/solid and solid/solid interaction:
- version 3 or version 4 of the Chrono library are supported;
- GPUSPH can be compiled without Chrono support; moving objects with prescribed motion (e.g. paddles and gates) and no force-feedback are supported in this configuration;
an MPI implementation, for multi-node simulations:
- the minimum recommended versions are OpenMPI 1.8.4 or MVAPICH2 1.8; other MPI implementations may work too without GPUDirect support;
- GPUSPH can be compiled without MPI support; multiple GPUs on a single machine are still supported in this configuration.

Known issues:

some CUDA-aware MPI implementations do not support working with multiple GPUs per process;
- a possible workaround to use all GPUs on multi-GPU nodes is to run multiple processes per node;
hot-starting a simulation with moving objects may produce slightly different results.

A discussion forum for GPUSPH developers and users is available to share ideas on GPUSPH and its applications. For technical issues, please use the official issue tracker. We also invite you to join the #gpusph IRC channel on FreeNode.

The GPUSPH Discourse group is now available

Mon, 18 Mar 2019 15:00:00 +0100

In our effort to improve the opportunities of collaboration within the GPUSPH community, we have set up a Discourse forum dedicated to GPUSPH.

We invite you all to join us there as the primary discussion forum for everything pertaining GPUSPH application and development: interesting test cases, tips & tricks on the GPUSPH usage, proposals about the user interface, discussions about the project direction, information about its development are all appropriate topics. The GitHub issue pages remain the preferred way to report specific bugs and feature requests.

We wish to thank Civilized Discourse Construction Kit, Inc. (the company developing the Discourse platform) for the generous offer to host our forum free of charge.

The GPUSPH development branch is now public

Thu, 08 Nov 2018 13:00:00 +0100

We are pleased to announce that the main GPUSPH development branch (next) has been made public, complementing the master (holding the current released version) and candidate (holding the feature-complete, in-testing release candidate for the upcoming version) branches.

This has been done in an effort to improve opportunities of collaboration with the community, as the GPUSPH code base is undergoing significant internal restructuring in preparation for version 5.

For more information about collaborating to the GPUSPH project, please consult CONTRIBUTING on the GPUSPH repository, which will be kept up to date as we improve the infrastructure for discussion and code sharing.

For any feedback or support request please use the official issue tracker.

We also invite you to join the #gpusph IRC channel on FreeNode.

GPUSPH v4.1 released

Wed, 19 Jul 2017 10:00:00 +0200

GPUSPH version 4.1 is out! Head over to the release tag on GitHub.

We are pleased to announce the official release of GPUSPH version 4.1. This new version introduces several fixes and improvements to GPUSPH over version 4.0:

open boundaries now have multi-GPU/multi-node support (only available for semi-analytical boundaries) (fixes GitHub issue #18);
multi-device simulations do not lock unused devices (fixes GitHub issue #17);
support for updated Chrono library (version 3.0) (should fix GitHub issue #19);
automatic water depth computation at open boundaries is now possible for all types of open boundaries (with prescribed pressure and prescribed velocity)
the hotstart (resume) functionality now works for simulations involving moving bodies (but see Known Issues below);
improved balanced domain split with SA boundaries.

Requirements:

Linux or Mac OS X;
- full support for Mac OS X may be phased out in the future, due to the difficulty in keeping up with the evolution of the CUDA and compiler support on that platform;
CUDA 7.0 or later;
a host compiler with C++11 support, and supported by CUDA:
- on Linux, GCC 4.8 or later, or Clang 3.6 or later (CUDA 7.5 or later only);
- on Mac OS X, XCode 6 or later;
the Chrono library is required for fluid/solid and solid/solid interaction;
- GPUSPH can be compiled without Chrono support;
- you need version 3.0 of the Chrono library. The develop branch of Chrono at the time of this announcement (7f3d61f641c41b63f58c79dc0c0355768864f911) is known to work too;
an MPI implementation for multi-node support;
- OpenMPI 1.8.4 or MVAPICH2 1.8 recommended, other MPI implementations may work too without GPUDirect support;
- GPUSPH can be compiled without MPI support.

Known issues:

large simulations may lock up on Maxwell class GPUs (Compute Capability 5.x, i.e. GTX 9xx, GTX Titan, Titan X) with CUDA version 7.x;
- this is actually an issue in the thrust library, which we use in the particle sort;
- upgrading to the latest version of CUDA fixes the issue (CUDA 8 at the time of writing);
hot-starting a simulation with moving objects may produce slightly different results;
simulations with moving objects running on multiple nodes may fail to resume correctly on hot-start;
if a simulation runs to completion, it is possible to resume from the last checkpoint and continue, but the results may be different from the ones that would have been obtained running the original simulation for longer;
Ferrari correction may not work correctly with boundary formulations different from semi-analytical;
the time step size may vary when using semi-analytical boundaries, even if a fixed time step is asked for.

For any feedback or support request please use the official issue tracker.

We also invite you to join the #gpusph IRC channel on FreeNode.

GPUSPH v4 released

Wed, 19 Oct 2016 15:00:00 +0200

GPUSPH version 4.0 is out! Head over to the release tag on GitHub.

We are pleased to announce the official release of GPUSPH version 4.0. After two years of development, the new version introduces significant computational advancements, new SPH modelling features, and a simpler and more robust interface to define simulation scenarios.

New features:

Gaussian smoothing kernel;
Grenier formulation for multi-fluid support;
dynamic particles boundary formulation;
improved support for periodic boundary conditions;
open boundaries (inlet/outlet) with the semi-analytical boundary formulation;
computation of total mechanical energy;
hot-start (possibility to restart a simulation from a previous run);
multi-node support: running a simulation across multiple cluster nodes, with one or more GPUs on each node;
new, simplified interface for the definition of simulation scenarios;
use of the Chrono library instead of ODE for moving objects (support for complex joints);
better, updated documentation.

Requirements:

Linux or Mac OS X;
- full support for Mac OS X may be phased out in the future, due to the difficulty in keeping up with the evolution of the CUDA and compiler support on that platform;
CUDA 7.0 or later;
a host compiler with C++11 support, and supported by CUDA:
- on Linux, GCC 4.8 or later, or Clang 3.6 or later (CUDA 7.5 or later only);
- on Mac OS X, XCode 6 or later;
the develop branch of the Chrono library is required for fluid/solid and solid/solid interaction;
- GPUSPH can be compiled without Chrono support;
- the latest Chrono commit at the moment of this announcement (984f51dbe16903097fe0170b4aabefa851421b63) is known to work;
an MPI implementation for multi-node support;
- OpenMPI 1.8.4 or MVAPICH2 1.8 recommended, other MPI implementations may work too without GPUDirect support;
- GPUSPH can be compiled without MPI support.

Known issues:

large simulations may lock up on Maxwell class GPUs (Compute Capability 5.x, i.e. GTX 9xx, GTX Titan, Titan X)
- this is actually an issue in the thrust library, which we use in the particle sort;
open boundaries do not work correctly in multi-GPU simulations;
GPUSPH may fail to run on systems with multiple GPUs when the first GPU is set in exclusive mode;
- this is due to GPUSPH always creating a context on the first GPU when pinning memory to speed up host/device data transfers;
- a fix is under development; as a temporary workaround, change the first GPU to non-exclusive mode;
some CUDA-aware MPI implementations do not support working with multiple GPUs per process;
- a possible workaround to use all GPUs on multi-GPU nodes is to run multiple processes per node;
hot-starting a simulation with moving objects may produce slightly different results;
Ferrari correction may not work correctly with boundary formulations different from semi-analytical;
the time step size may vary when using semi-analytical boundaries, even if a fixed time step is asked for.

For any feedback or support request please use the official issue tracker.

We also invite you to join the #gpusph IRC channel on FreeNode.

GPUSPH v3 released

Tue, 03 Jun 2014 09:00:00 +0200

GPUSPH version 3.0 is out! Head over to the release tag on GitHub.

GPUSPH v3.0 has been released during the 9th International SPHERIC Workshop, and includes the new features (homogeneous accuracy, multi-GPU and multi-node, fully coupled fluid/object interaction, unified semi-analytical boundaries) presented at the conference.

New GPUSPH website

Sat, 31 May 2014 01:13:56 +0200

With the upcoming 3.0 release of GPUSPH, the first implementation of weakly-compressible SPH to run entirely on CUDA GPUs, we decided to design a new website for your viewing pleasure.

We intend to keep you posted on the newest developments of GPUSPH. Feel free to peruse the list of publications, enjoy the gallery, or go grab the source on our public GitHub repository.