InfoQ

News

MPI for .NET

Posted by Jonathan Allen on Jan 23, 2008 11:20 AM

Community
.NET
Topics
Performance & Scalability
Tags
Distributed Programming,
Parallel Programming

MPI or Message Passing Interface is the standard for distributed programming such as that used in supercomputers and implementations can be found for FORTRAN, C, and C++. There are several projects in the works to bring that power to .NET. Today we look at two of them.

The first is Pure Mpi.NET. This implementation is written entirely in managed code and heavily leverages the Windows Communication Foundation. It also changes the API to better match the .NET style of framework writing, leaving one to wonder if there is anything left of MPI in Pure Mpi.NET.

Pure Mpi.NET has another flaw; its licensing agreement is rather onerous. The license is only granted to non-profit organizations such as schools and individuals. For profit companies are explicitly from using the library.

The next is MPI.NET. This is an open source project based on Boost.MPI. Though currently a technology preview, the API is already well documented. In addition to not being production grade, MPI.NET is tied to Microsoft's MPI implementation, MS-MPI. This means that production systems will have to run Microsoft Compute Cluster Server 2003 at 469 USD/server.

When one looks at the sorry state of MPI support on the .NET platform, it is hard not to wonder why it has taken so long to catch up to languages like FORTRAN for distributed programming.

Related Sponsor

Terracotta is Scalability and Availability for Java Applications. It clusters the JVM itself, which dramatically simplifies development and reduces database dependency.

1 comment

Reply

MPI for .NET by the dude Posted Jan 25, 2008 12:33 PM

  1. Back to top

    MPI for .NET

    Jan 25, 2008 12:33 PM by the dude

    Catch up to FORTRAN? Are you serious? FORTRAN has been dead for a looong time. .NET is here, it's not going anywhere, and is the platform of the future. It's great for distributed systems, but lacks some things for distributed parallel programming. It doesn't have a single program / multiple data parallel model, and these implementations fill in that gap.

    Reply

Exclusive Content

Agile Project Management: Lessons Learned at Google

In this presentation filmed during QCon 2007, Jeff Sutherland, the creator of Scrum, talks about his visit at Google to do an analysis of Google's first implementation of Scrum.

AtomServer – The Power of Publishing for Data Distribution

In this article, Bryon Jacob and Chris Berry introduce AtomServer, their implementation of a full-fledged Atom Store based on Apache Abdera, which is now available as open source.

An Introduction to Virtualization

It is easy to think that virtualization applies only to servers. In reality the recent resurgence of the concept is also being applied to networking, storage, and application infrastructure.

REST Anti-Patterns

In this article, Stefan Tilkov explains some of the most common anti-patterns found in applications that claim to follow a "RESTful" design and suggests ways to avoid them.

Choosing between Routing and Orchestration in an ESB

In this article, Adrien Louis and Marc Dutoo discuss the differences and relative merits of using orchestration vs. routing in a typical ESB setup, and discuss various implementation options.

Enterprise Batch Processing with Spring

Wayne Lund discusses batch processing, Spring Batch objectives and features, scenarios for usage, Spring Batch architecture, scaling, example code, failures and retrying, and the future roadmap.

User Story Estimation Techniques

Developer Jay Fields draws on his experiences as a ThoughtWorks consultant to describe effective user story estimation techniques.

Security (CAS and OpenID) with Ruby

In this talk from QCon SF 2007, Justin Gehtland explains two open solutions to distributed identity and their Rails integration components: OpenID (using ruby-openid) and CAS (using rubycas-client).