MPI#

Source code checked in, #94877

Project Collection Service Accounts — Mon, 01 Oct 2012 22:38:35 GMT

Upgrade: New Version of LabDefaultTemplate.xaml. To upgrade your build definitions, please visit the following link: http://go.microsoft.com/fwlink/?LinkId=254563

Source code checked in, #94876

Project Collection Service Accounts — Mon, 01 Oct 2012 22:31:06 GMT

Checked in by server upgrade

Source code checked in, #57274

_TFSSERVICE — Wed, 28 Jul 2010 22:13:23 GMT

Checked in by server upgrade

Updated Wiki: Home

faafonso — Wed, 28 Apr 2010 13:21:48 GMT

Equipe/Team | Blog

Project Status
First Steps to Assembly MPI.NET Aplications

Our next release will be available soon. It's a very faster version of the Spawn method.

We are testing a release of a new version of the MPI_Comm_Spawn primitive witch is very faster and allows the spawn of methods.

The implementation of the MPI_Comm_Spawn_multiple is now available at the downloads page http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=30261

The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented and available for download at http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25184

We are now entering the second part of our project. See the Project Status page for more details.

Project Description

MPI is the de-facto standard for HPC. The 2.0 standard specifies bindings for Fortran, C and C. However, they clearly lack of high-level abstractions such are OO encapsulation or simple portability in heterogeneous platforms, i.e. distributed sets of different CPUs, possibly interconnected by different networks.

Nearly all current, modern languages, have developed extensions such as to enable their use with MPI: this is the case of Java (JavaMPI, MPIJava), Python (pympi), Perl or Ruby. As far as C# is concerned, the Open Systems Laboratory at Indiana University has proposed both a low-level binding (strongly inspired by the C binding) and a high-level one, called MPI.NET (see Willcock, J., Lumsdaine, A., Robison., A.: Using mpi with c# and the common language infrastructure. Concurrency and Computation: Practice and Experience 17(7-8) (2005) 895–917).

The C# binding is relatively straightforward. Each object of C# bindings contains the underlying C representation of the MPI object. Similarly, the high-level objects in the MPI.NET are usually containers of underlying MPI objects. According to the referenced article, the performance of the current MPI binding of C# is reasonnable. However, the implementation and the tests have only be partial and did not cover collective communication, one of the key features of MPI, neither do they cover non-blocking communication or other advanced features of MPI such as the use of non-native, used-defined datatypes.

The goals of this project would be to build upon MPI.NET in order to complement it with the features that are missing, mainly regarding collective communication. Either they could benefit from C# native support for such communication, either they could be programmed on top of the provided MPISend/MPIRecv encapsulations. C# and .NET features such as fault tolerance or dynamicity support would be studied, in other to turn the MPI# implementation robust in large, dynamic and heterogeneous platforms.

Updated Release: MPI.NET Comm_spawn for Windows (mar 31, 2009)

faafonso — Tue, 22 Dec 2009 03:28:29 GMT

2009/12/1 - Please be aware that in windows this library still in a beta version.

This release has the MPI_Comm_spawn and the MPI_Comm_get_parent primitives implemented on the MPI.NET library.
The primitives are working and has their interfaces at the Communicator class in the methods Communicator.Spawn and Communicator.GetParent.
I modified the Hello and the Hostname applications to test the implementation.
The platform used was Windows XP with MPICH2 and MPI.NET 1.0.0.
Suggestions and corrections are welcome.
Thanks,
Fernando.

Released: MPI.NET Comm_spawn for Windows (Mar 31, 2009)

Tue, 22 Dec 2009 03:28:29 GMT

Updated Release: MPI.NET Comm_spawn for Linux (mar 24, 2009)

faafonso — Tue, 22 Dec 2009 03:27:11 GMT

2009/12/1 - We are recommending the use of mpich2 1.2.1 for better performance and stability. The library is being modified and soon a new release should be available.

2009/04/16 - A few modifications where made at the library and I added some new function overriding. I also tested the library with mpich2. I tested the library with the Fibonacci application and got better stability results with mpich2 than with openmpi for both pure c and cpp mpi and mpi.net mpi. The mpich2 version used is the 1.0.8.

This release has the MPI_Comm_spawn and the MPI_Comm_get_parent primitives implemented on the MPI.NET library.
The primitives are working and has their interfaces at the Communicator class in the methods Communicator.Spawn and Communicator.GetParent.
I modified the Hello and the Hostname applications to test the implementation.
The platform used was Ubuntu Linux 8.10 with OpenMPI 1.3 and MPI.NET 1.0.0.
Suggestions and corrections are welcome.
Thanks,
Fernando.

REALLY IMPORTANT NOTE:
Please use autoconf-2.61, automake-1.10.1 and libtool-2.2.4 versions.
They can be downloaded at ftp://ftp.gnu.org/gnu/

Released: MPI.NET Comm_spawn for Linux (Mar 24, 2009)

Tue, 22 Dec 2009 03:27:11 GMT

Updated Wiki: Home

faafonso — Tue, 22 Dec 2009 03:22:00 GMT

Equipe/Team | Blog

Project Status
First Steps to Assembly MPI.NET Aplications

We are testing a release of a new version of the MPI_Comm_Spawn primitive witch is very faster and allows the spawn of methods.

The implementation of the MPI_Comm_Spawn_multiple is now available at the downloads page http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=30261

The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented and available for download at http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25184

We are now entering the second part of our project. See the Project Status page for more details.

Project Description

MPI is the de-facto standard for HPC. The 2.0 standard specifies bindings for Fortran, C and C. However, they clearly lack of high-level abstractions such are OO encapsulation or simple portability in heterogeneous platforms, i.e. distributed sets of different CPUs, possibly interconnected by different networks.

Nearly all current, modern languages, have developed extensions such as to enable their use with MPI: this is the case of Java (JavaMPI, MPIJava), Python (pympi), Perl or Ruby. As far as C# is concerned, the Open Systems Laboratory at Indiana University has proposed both a low-level binding (strongly inspired by the C binding) and a high-level one, called MPI.NET (see Willcock, J., Lumsdaine, A., Robison., A.: Using mpi with c# and the common language infrastructure. Concurrency and Computation: Practice and Experience 17(7-8) (2005) 895–917).

The C# binding is relatively straightforward. Each object of C# bindings contains the underlying C representation of the MPI object. Similarly, the high-level objects in the MPI.NET are usually containers of underlying MPI objects. According to the referenced article, the performance of the current MPI binding of C# is reasonnable. However, the implementation and the tests have only be partial and did not cover collective communication, one of the key features of MPI, neither do they cover non-blocking communication or other advanced features of MPI such as the use of non-native, used-defined datatypes.

The goals of this project would be to build upon MPI.NET in order to complement it with the features that are missing, mainly regarding collective communication. Either they could benefit from C# native support for such communication, either they could be programmed on top of the provided MPISend/MPIRecv encapsulations. C# and .NET features such as fault tolerance or dynamicity support would be studied, in other to turn the MPI# implementation robust in large, dynamic and heterogeneous platforms.

Updated Wiki: Home

faafonso — Tue, 22 Dec 2009 03:11:29 GMT

Updated Wiki: Project Status

faafonso — Tue, 22 Dec 2009 03:09:40 GMT

First part of the project

2009/12/01
- We are running benchmarks over the new and faster MPI.NET-Spawn library. See the Fibonacci performance graph below.

2009/10/07
- We are working on a faster version of the Spawn primitive.

2009/10/07
- We published a papper in a journal witch is available at http://www.unisinos.br/publicacoes_cientificas/images/stories/Publicacoes/Scientia_v20n1/53-64.pdf.

2009/07/03
- We are running some benchmarks over the MPI_Comm_Spawn mechanism. We are also working on some pappers.

2009/03/24
- The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented (see the releases).

2009/02/20
- The MPI_Comm_Spawn primitive implementation worked, so now I'm focusing on improve the implementation and make it's interface looks nice. The preliminary implementation can be found on the Releases.

2009/01/02
- The MPI_Comm_Spawn primitive implementation started.

Updated Wiki: Equipe/Team

faafonso — Tue, 22 Dec 2009 02:59:30 GMT

Fernando Abrahão Afonso

Updated Wiki: Home

faafonso — Wed, 07 Oct 2009 19:51:39 GMT

Equipe/Team | Blog

Project Status
First Steps to Assembly MPI.NET Aplications

We are planning a release of a new version of the MPI_Comm_Spawn primitive witch is very faster and allows the spawn of methods.

The implementation of the MPI_Comm_Spawn_multiple is now available at the downloads page http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=30261

The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented and available for download at http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25184

We are now entering the second part of our project. See the Project Status page for more details.

Project Description

MPI is the de-facto standard for HPC. The 2.0 standard specifies bindings for Fortran, C and C. However, they clearly lack of high-level abstractions such are OO encapsulation or simple portability in heterogeneous platforms, i.e. distributed sets of different CPUs, possibly interconnected by different networks.

Nearly all current, modern languages, have developed extensions such as to enable their use with MPI: this is the case of Java (JavaMPI, MPIJava), Python (pympi), Perl or Ruby. As far as C# is concerned, the Open Systems Laboratory at Indiana University has proposed both a low-level binding (strongly inspired by the C binding) and a high-level one, called MPI.NET (see Willcock, J., Lumsdaine, A., Robison., A.: Using mpi with c# and the common language infrastructure. Concurrency and Computation: Practice and Experience 17(7-8) (2005) 895–917).

The C# binding is relatively straightforward. Each object of C# bindings contains the underlying C representation of the MPI object. Similarly, the high-level objects in the MPI.NET are usually containers of underlying MPI objects. According to the referenced article, the performance of the current MPI binding of C# is reasonnable. However, the implementation and the tests have only be partial and did not cover collective communication, one of the key features of MPI, neither do they cover non-blocking communication or other advanced features of MPI such as the use of non-native, used-defined datatypes.

The goals of this project would be to build upon MPI.NET in order to complement it with the features that are missing, mainly regarding collective communication. Either they could benefit from C# native support for such communication, either they could be programmed on top of the provided MPISend/MPIRecv encapsulations. C# and .NET features such as fault tolerance or dynamicity support would be studied, in other to turn the MPI# implementation robust in large, dynamic and heterogeneous platforms.

Status do Projeto

Descrição do Projeto

O objetivo do projeto MPI# é construir, acima do ambiente MPI.NET, os complementos que faltam para permitir a comunicação entre processos em ambientes de alta performance construídos na linguagem C#, de forma nativa. Aspectos de tolerância a falha e suporte dinâmico dos ambientes C# e .Net serão estudados, com o intuito de permitir a implementação do MPI# robusta em ambientes com múltiplos processadores, dinâmicos e com plataformas heterogêneas.

Updated Wiki: Project Status

faafonso — Wed, 07 Oct 2009 19:50:09 GMT

First part of the project

2009/10/07
- We are working on a faster version of the Spawn primitive.

2009/10/07
- We published a papper in a journal witch is available at http://www.unisinos.br/publicacoes_cientificas/images/stories/Publicacoes/Scientia_v20n1/53-64.pdf.

2009/07/03
- We are running some benchmarks over the MPI_Comm_Spawn mechanism. We are also working on some pappers.

2009/03/24
- The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented (see the releases).

2009/02/20
- The MPI_Comm_Spawn primitive implementation worked, so now I'm focusing on improve the implementation and make it's interface looks nice. The preliminary implementation can be found on the Releases.

2009/01/02
- The MPI_Comm_Spawn primitive implementation started.

Updated Wiki: Project Status

faafonso — Wed, 07 Oct 2009 19:48:41 GMT

First part of the project

2009/10/07
- We published a papper in a journal witch is available at http://www.unisinos.br/publicacoes_cientificas/images/stories/Publicacoes/Scientia_v20n1/53-64.pdf.

2009/07/03
- We are running some benchmarks over the MPI_Comm_Spawn mechanism. We are also working on some pappers.

2009/03/24
- The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented (see the releases).

2009/02/20
- The MPI_Comm_Spawn primitive implementation worked, so now I'm focusing on improve the implementation and make it's interface looks nice. The preliminary implementation can be found on the Releases.

2009/01/02
- The MPI_Comm_Spawn primitive implementation started.

Updated Wiki: Project Status

faafonso — Wed, 07 Oct 2009 19:48:08 GMT

First part of the project

2009/10/07
- We published a papper in a journal witch is available at http://www.unisinos.br/publicacoes_cientificas/images/stories/Publicacoes/Scientia_v20n1/53-64.pdf.

2009/07/03
- We are running some benchmarks over the MPI_Comm_Spawn mechanism. We are also working on some pappers.

2009/03/24
- The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented (see the releases).

2009/02/20
- The MPI_Comm_Spawn primitive implementation worked, so now I'm focusing on improve the implementation and make it's interface looks nice. The preliminary implementation can be found on the Releases.

2009/01/02
- The MPI_Comm_Spawn primitive implementation started.

Updated Release: MPI.NET Comm_spawn for Linux (mar 24, 2009)

faafonso — Thu, 27 Aug 2009 19:31:05 GMT

2009/04/16 - A few modifications where made at the library and I added some new function overriding. I also tested the library with mpich2. I tested the library with the Fibonacci application and got better stability results with mpich2 than with openmpi for both pure c and cpp mpi and mpi.net mpi. The mpich2 version used is the 1.0.8.

This release has the MPI_Comm_spawn and the MPI_Comm_get_parent primitives implemented on the MPI.NET library.
The primitives are working and has their interfaces at the Communicator class in the methods Communicator.Spawn and Communicator.GetParent.
I modified the Hello and the Hostname applications to test the implementation.
The platform used was Ubuntu Linux 8.10 with OpenMPI 1.3 and MPI.NET 1.0.0.
Suggestions and corrections are welcome.
Thanks,
Fernando.

REALLY IMPORTANT NOTE:
Please use autoconf-2.61, automake-1.10.1 and libtool-2.2.4 versions.
They can be downloaded at ftp://ftp.gnu.org/gnu/

Released: MPI.NET Comm_spawn for Linux (Mar 24, 2009)

Thu, 27 Aug 2009 19:31:05 GMT

Updated Wiki: Home

faafonso — Thu, 16 Jul 2009 20:19:59 GMT

Equipe/Team | Blog

Project Status
First Steps to Assembly MPI.NET Aplications

The implementation of the MPI_Comm_Spawn_multiple is now available at the downloads page http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=30261

The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented and available for download at http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25184

We are now entering the second part of our project. See the Project Status page for more details.

Project Description

MPI is the de-facto standard for HPC. The 2.0 standard specifies bindings for Fortran, C and C. However, they clearly lack of high-level abstractions such are OO encapsulation or simple portability in heterogeneous platforms, i.e. distributed sets of different CPUs, possibly interconnected by different networks.

Nearly all current, modern languages, have developed extensions such as to enable their use with MPI: this is the case of Java (JavaMPI, MPIJava), Python (pympi), Perl or Ruby. As far as C# is concerned, the Open Systems Laboratory at Indiana University has proposed both a low-level binding (strongly inspired by the C binding) and a high-level one, called MPI.NET (see Willcock, J., Lumsdaine, A., Robison., A.: Using mpi with c# and the common language infrastructure. Concurrency and Computation: Practice and Experience 17(7-8) (2005) 895–917).

The C# binding is relatively straightforward. Each object of C# bindings contains the underlying C representation of the MPI object. Similarly, the high-level objects in the MPI.NET are usually containers of underlying MPI objects. According to the referenced article, the performance of the current MPI binding of C# is reasonnable. However, the implementation and the tests have only be partial and did not cover collective communication, one of the key features of MPI, neither do they cover non-blocking communication or other advanced features of MPI such as the use of non-native, used-defined datatypes.

The goals of this project would be to build upon MPI.NET in order to complement it with the features that are missing, mainly regarding collective communication. Either they could benefit from C# native support for such communication, either they could be programmed on top of the provided MPISend/MPIRecv encapsulations. C# and .NET features such as fault tolerance or dynamicity support would be studied, in other to turn the MPI# implementation robust in large, dynamic and heterogeneous platforms.

Status do Projeto

Descrição do Projeto

O objetivo do projeto MPI# é construir, acima do ambiente MPI.NET, os complementos que faltam para permitir a comunicação entre processos em ambientes de alta performance construídos na linguagem C#, de forma nativa. Aspectos de tolerância a falha e suporte dinâmico dos ambientes C# e .Net serão estudados, com o intuito de permitir a implementação do MPI# robusta em ambientes com múltiplos processadores, dinâmicos e com plataformas heterogêneas.

Updated Wiki: Home

faafonso — Thu, 16 Jul 2009 20:18:01 GMT

Equipe/Team | Blog

Project Status
First Steps to Assembly MPI.NET Aplications

The MPI_Comm_Spawn and MPI_Comm_get_parent primitives are implemented and available for download at http://mpihash.codeplex.com/Release/ProjectReleases.aspx?ReleaseId=25184*

We are now entering the second part of our project. See the Project Status page for more details.

Project Description

MPI is the de-facto standard for HPC. The 2.0 standard specifies bindings for Fortran, C and C. However, they clearly lack of high-level abstractions such are OO encapsulation or simple portability in heterogeneous platforms, i.e. distributed sets of different CPUs, possibly interconnected by different networks.

Nearly all current, modern languages, have developed extensions such as to enable their use with MPI: this is the case of Java (JavaMPI, MPIJava), Python (pympi), Perl or Ruby. As far as C# is concerned, the Open Systems Laboratory at Indiana University has proposed both a low-level binding (strongly inspired by the C binding) and a high-level one, called MPI.NET (see Willcock, J., Lumsdaine, A., Robison., A.: Using mpi with c# and the common language infrastructure. Concurrency and Computation: Practice and Experience 17(7-8) (2005) 895–917).

The C# binding is relatively straightforward. Each object of C# bindings contains the underlying C representation of the MPI object. Similarly, the high-level objects in the MPI.NET are usually containers of underlying MPI objects. According to the referenced article, the performance of the current MPI binding of C# is reasonnable. However, the implementation and the tests have only be partial and did not cover collective communication, one of the key features of MPI, neither do they cover non-blocking communication or other advanced features of MPI such as the use of non-native, used-defined datatypes.

The goals of this project would be to build upon MPI.NET in order to complement it with the features that are missing, mainly regarding collective communication. Either they could benefit from C# native support for such communication, either they could be programmed on top of the provided MPISend/MPIRecv encapsulations. C# and .NET features such as fault tolerance or dynamicity support would be studied, in other to turn the MPI# implementation robust in large, dynamic and heterogeneous platforms.

Status do Projeto

Descrição do Projeto

O objetivo do projeto MPI# é construir, acima do ambiente MPI.NET, os complementos que faltam para permitir a comunicação entre processos em ambientes de alta performance construídos na linguagem C#, de forma nativa. Aspectos de tolerância a falha e suporte dinâmico dos ambientes C# e .Net serão estudados, com o intuito de permitir a implementação do MPI# robusta em ambientes com múltiplos processadores, dinâmicos e com plataformas heterogêneas.