eScience @ Microsoft : WinHPC

TAU – perf tuning for HPC

eScience — Wed, 25 Nov 2009 21:03:00 GMT

TAU, the performance tuning OSS software for HPC on Linux has been successfully made available to support Microsoft Windows 7, Microsoft Windows HPC Server 2008. Please see release note section 5.

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Cray Brings Windows 7 to HPC

eScience — Fri, 13 Nov 2009 03:41:09 GMT

Now what computational scientist wouldn’t want one of these under their desk. I like the tagline, “putting ‘work’ back into workstation”.

Cray Brings Windows 7 to HPC

by Michael Feldman, HPCwire Editor

If you thought Windows 7 was just for mere mortals, think again. Microsoft's latest OS is about to show up in Cray's newest CX1 deskside supercomputer that puts a Windows workstation and a Windows HPC Server cluster into a single box. Called the CX1-iWS (for integrated workstation), the machines are to be sold exclusively through Dell and will range in price from $39K to $55K.

The idea behind the iWS is to retain the interactive experience of a personal workstation, but extend its computational power to that of a small HPC cluster. It's generally aimed at technical computing users who have simply run out of compute headroom on their two-socket machines, but are loathe to give up the intimacy of the workstation environment. A generic CX1 can be configured to provide the equivalent capabilities, but the iWS is preconfigured to deliver this experience right out of the box.

Since Windows 7 and Windows HPC Server 2008 form a natural client-server relationship, cluster administration and job management becomes relatively seamless. In addition, since there is disk storage shared between the workstation and the cluster, data management becomes much more straightforward. As long as your data set fits in 4 TB, no data transfers back and forth between client and server will be necessary.

HPCwire: Cray Brings Windows 7 to HPC

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Graywulf Takes Byte Out of Data Overload

eScience — Fri, 08 May 2009 20:46:17 GMT

Graywulf is the natural evolution of Beowulf Clusters – it brings together HPC clusters and databases to do efficient processing and data management. It’s name and design also pays homage to Jim Gray – who helped champion the use of relational databases in the scientific projects.

At it’s simplest form Graywulf is having a database installed on each of the HPC compute nodes – this brings the data to the computation – one of the points Jim made quite often and utilizes the power of databases (queries, stored procedures, etc). Since it’s a generic architecture Graywulf clusters can be built using any OS and any database…the ones in the case study below implemented them using Windows HPC Server and SQL Server and the motivation was to be more efficient in doing the science – it’s always great to have innovative folks using technologies to do good work.

“To put it simply, a scientist needs to be able to live within the data,” says Alexander Szalay, a cosmologist-turned-computer-scientist at The Johns Hopkins University (JHU) in Baltimore, Maryland. The power of information, Szalay says, is determined not by its quantity so much as how easy it is to access, manipulate and analyze.
“It’s not just about doing the numerical calculations,” adds Andrew Simms, a biomedical health informatics graduate student working on protein structure analysis in Valerie Daggett’s bioengineering laboratory at the University of Washington (UW) in Seattle. “It’s also about assembling the data so we can run calculations while performing analyses and ad hoc explorations and then feed it all back into the data warehouse.”

Graywulf Takes Byte Out of Data Overload

Astronomers at The Johns Hopkins University and protein scientists at the University of Washington are using inexpensive computer hardware combined with powerful computing and database software to help manage and analyze a growing volume of scientific data.

For details, read the Graywulf case study.

Project Principals

Alexander Szalay, Alumni Centennial Professor, Department of Physics and Astronomy, The Johns Hopkins University

Valerie Daggett, Professor of Bioengineering, University of Washington

Graywulf Takes Byte Out of Data Overload - Microsoft Research

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Windows HPC Server 2008 cracks Top 10 in Top 500 list

eScience — Tue, 18 Nov 2008 08:34:59 GMT

The latest Top 500 list was released at Supercomputing Conference (SC08) in Austin, Texas today and Windows HPC Server 2008 was part of the #10 placed machine [Dawning 5000A] from Shanghai Supercomputer Center. This is the top machine outside of US – amazing job by the Windows HPC team.

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

HPC Server 2008 Experience Counts Academic Program

eScience — Tue, 11 Nov 2008 00:49:34 GMT

Great opportunity to get a special academic discount on Windows HPC Server 2008.

Windows® HPC Server 2008 Experience Counts Academic Program

You know the value of high-performance computing (HPC). No matter the topic - the inner workings of the universe or the economy; the spawning of a hurricane or urban development - HPC has become a key driver of the type of research that leads to groundbreaking insights.

Microsoft Higher Education - Experience Counts

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

HPC - MPI.NET 1.0 Is Now Released...!

eScience — Wed, 08 Oct 2008 21:12:04 GMT

PhilPen has posted that the folks at Indiana University have released MPI.NET: High Performance C# library for Message Passing. The runtime and source code are available for download as well as a Tutorial. This should make it much easier for folks to use any .Net language to write MPI apps – I’m interested in seeing ones written with F# and even PowerShell.

MPI.NET 1.0 Is Now Released...!

MPI.NET is a high-performance, easy-to-use implementation of the Message Passing Interface (MPI) for Microsoft's .NET environment. MPI is the de facto standard for writing parallel programs running on a distributed memory system, such as a compute cluster, and is widely implemented. Most MPI implementations provide support for writing MPI programs in C, C++, and Fortran. MPI.NET provides support for all of the .NET languages (especially C#), and includes significant extensions (such as automatic serialization of objects) that make it far easier to build parallel programs that run on clusters.

MPI.NET has been developed by the research staff at Indiana University in collaboration with Microsoft. Developers leverage the "Windows HPC Server 2008 SDK" in tandem with the MPI.NET SDK to build MPI.NET applications. An MPI.NET runtime component must be installed onto Windows HPC Server 2008 based clusters to host MPI.NET applications.

Regarding Windows Server : MPI.NET 1.0 Is Now Released...!

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Windows HPC and Pixar animation software RenderMan

eScience — Wed, 20 Aug 2008 09:49:47 GMT

It’s great to see that Pixar will be releasing RenderMan on the upcoming release of Windows HPC 2008.

The rendering software behind such hit animated movies as The Incredibles, Ratatouille and Monsters Inc. should become even more powerful when it becomes available on Microsoft Corp.'s most advanced operating systems later this year.

At the Siggraph 2008 computer graphics conference in Los Angeles this week, Pixar Animation Studios Inc. said that a new RenderMan Pro Server 14.0 release will be the first version of the software that runs on Windows HPC Server 2008, the upcoming release of Microsoft's operating system for high-performance computing clusters and supercomputers. RenderMan 14.0 will also support the 64-bit version of Windows Vista, according to Pixar's announcement.

Siggraph: Pixar animation software to get boost on Windows supercomputers

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

HPC 2008 & NCSA – 23rd on Top500 – Great Efficiency

eScience — Wed, 18 Jun 2008 21:01:30 GMT

The Top500 list was released and the cluster at NCSA came in at 23rd running a beta of Windows HPC 2008. What’s really interesting is the 77% efficiency and that it only took 4 hours to install the 1200 node cluster.

Press Article - Windows HPC Server 2008 debuts in top 25 of the world’s TOP500 largest supercomputers. Ranked at 23, the NCSA cluster performed at 68.5 TFlops and 77.7% efficiency.

Video

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Windows HPC Server 2008 Beta 2 is Here

eScience — Fri, 23 May 2008 18:32:47 GMT

Windows HPC Server 2008 Beta 2 is available on Microsoft Connect and Ryan Waite goes into detail on much of the new stuff. I'm really excited to start seeing scientists utilize and evaluate the new version - I think many of the new pieces especially Network Direct, PowerShell integration, and the HPC Basic Profile will surprise users. The other piece I'm looking forward to is the release of the next Top500 list in June...

We completed a few great Top500 runs in the last few weeks. We can’t talk about the numbers until the International Supercomputing Conference in June but it looks like Beta 2’s new MPI stack and new Network Direct RDMA interface are starting to hum.
Windows HPC Server 2008 Beta 2 is Here

HPC Server 2008 Beta Technical Overview

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Windows HPC - Computational Finance Pilot

eScience — Fri, 28 Mar 2008 19:16:06 GMT

The Windows HPC team has just made public their Computational Finance Pilot where they are enabling execution of computational fiance models for university courses - there is a good paper on the implementation. It would be good to see if the same type of implementation could be used for offering up science based services. The pilot brings is comprised of the following components:

Computing Resources - A 64-node\256-core compute cluster with 5 TB storage and a low latency interconnect
Central Market Dataset - Historical market data including 5 years of intraday equity tick data for the S&P 500, daily and fundamental data for 10,000 stocks and mortgage backed securities pool data
Microsoft® Office SharePoint® Server 2007 Web Portal - A Microsoft Office SharePoint Server 2007 portal to publish, browse and monitor models
Excel® HPC Task Pane - A Microsoft Excel 2007 user interface for model input and results
Model Execution Status Notifications - Model execution workflow with status email notifications

Introducing the Microsoft HPC++ CompFin Lab

The Microsoft HPC++ CompFin Lab integrates Microsoft HPC Server, a central market data database and Microsoft productivity products to provide university courses and research with an online service to publish, execute and manage computational finance models. Read more about it on our home page located here.

Windows HPC Community - CompFin Pilot

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Silverlight-based cycle stealing

eScience — Mon, 28 Jan 2008 23:12:20 GMT

Savas and I had talked about this idea of using Silverlight-based cycle stealing and wondered how well it would work. It's good to see this article on CodeProject about Legion: Build your own virtual super computer with Silverlight by Daniel Vaughan.

Legion is a Grid Computing framework that uses the Silverlight CLR to execute user definable tasks. Legion uses an ASP.NET application and web services to download tasks, upload result data, and provide grid-wide thread-safe operations for web clients or agents. Multiple tasks can be hosted at once, with Legion managing the delegation of tasks to agents. Client performance metrics, such as bandwidth and processor speed, may be used to tailor jobs for clients. Legion provides a management service and WPF application that is used to monitor the Legion grid.
I have deployed Legion to a demonstration server here so you can see it in action.

I wonder if Daniel is aware of the previous Legion grid system by Andrew Grimshaw that turned into Avaki (now part of Sybase)

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Windows HPC Server 2008 - 1 day training

eScience — Mon, 28 Jan 2008 21:16:51 GMT

Here's a good opportunity to get the latest information on the next version of the Windows Compute Cluster - now named Windows HPC Server 2008. If you're wondering why Windows HPC Server 2008 is a hot topic - take a look at the performance increases they've made and the technical overview.

HPC PAL

Microsoft invites you to participate in our HPCPAL Jumpstart technical training series for IT Pro's and Developers. This one day immersion style event will help you get the technical information you require to start deploying and developing HPC solutions leveraging Windows HPC Server 2008.
You will experience how Windows HPC Server 2008 platform is simple to deploy, operate, and integrate with existing infrastructure and tools.
1-Day Jumpstart

SPONSORS: Discuss clustered computing solution scenarios with Microsoft HPC technology specialists.
ROADMAP: Review the HPC platform technology roadmap for Windows HPC.
LEARN ABOUT: Microsoft Compute Cluster Pack v2 features and capabilities
EXPERIENCE: The significant advancements coming in the area of cluster management, scalability and performance
EXPLORE: HPC Development tools and techniques
SOLUTIONS: Identify integrated solution scenarios leveraging Windows platform technologies.
LEAVE WITH: Tools to equip your team to develop and deploy HPC solutions.

Check out at the dates at the HPC site

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

January update to Science @ Microsoft

eScience — Sat, 19 Jan 2008 11:02:52 GMT

The January update to the Technical Computing @ Microsoft site (www.microsoft.com/science) is up:

Spotlighting the SwissEx Collaboration with EPFL.
A feature entitled “Determining Fundamental Principles of RNA Structure with Comparative Sequence Analysis” about Dr. Robin Gutell (University of Texas at Austin) and the MSR collaboration (Stuart Ozer) around the use of SQL Server.
A highlight of the Microsoft Rainier Cluster using Windows HPC Server 2008 and how it’s more 30% more efficient.
Story about the American Geophysical Union’s conference in San Francisco last month and MSR contributions (Catharine Van Ingen) to the climate community.
Updates to the Books & Papers page, including the paper “A Virtual Research Environment for Bioscience Researchers”.

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

MPI.NET: High-Performance C# Library for Message Passing

eScience — Fri, 18 Jan 2008 04:11:36 GMT

It's great to see a high-performance .net library for writing MPI apps in C#. It would be good to hear about implementations using MPI.net and how it works. It would also be neat to see someone test out MPI.net with F#. Can't wait to see the published paper by Douglas Gregor and Andrew Lumsdaine Design and Implementation of a High-Performance MPI for C# and the Common Language Infrastructure that will appear in PPoPP 2008.

MPI.NET: High-Performance C# Library for Message Passing

MPI.NET is a high-performance, easy-to-use implementation of the Message Passing Interface (MPI) for Microsoft's .NET environment. MPI is the de facto standard for writing parallel programs running on a distributed memory system, such as a compute cluster, and is widely implemented. Most MPI implementations provide support for writing MPI programs in C, C++, and Fortran. MPI.NET provides support for all of the .NET languages (especially C#), and includes significant extensions (such as automatic serialization of objects) that make it far easier to build parallel programs that run on clusters.

MPI.NET: High-Performance C# Library for Message Passing

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)

Supercomputing 07

eScience — Thu, 22 Nov 2007 19:24:34 GMT

Was at Supercomputing last week - overall the event went really well. There was also a few announcements from the HPC team - the big one being the name and beta availability of Windows HPC Server 2008 and the new banners.

Attended part of the Manycore and Multicore Computing: Architectures, Applications And Directions Workshop, which had a packed house - I'm guessing due to both the topic as well as the stellar list of speakers.

In the Microsoft booth at SC - our collaborators from UCSD and Oxford had a optiputer demo featuring a large multi-tile display showing HD Streams of researchers at Oxford using a microscope at UCSD and analyzing the results on a a HPC Cluster at Oxford - all while collaborating using the display.

One highlights of the event - was an international man of mystery giving out awards to the HPC institutes..

Cross Posted from Dan Fay's Blog (http://blogs.msdn.com/dan_fay)