Heading to Hamburg for ISC’11? The HPC Advisory Council has published the agenda for their European Workshop event on Sunday, June 19. The workshop will focus on HPC productivity, and advanced HPC topics and futures, and will bring together system managers, researchers, developers, computational scientists and industry affiliates to discuss recent developments and future advancements in High-Performance Computing. Read the Full Story.
Start ISC’11 with HPC Advisory Council European Workshop June 19
Guest Feature: Looking forward to ISC 2011
In this special guest feature, Intel’s John Hengeveld lists the things he’ll be looking at ISC’11.
I love the now annual trek to Hamburg for ISC. I love the eclecticism of the city. I love the garden by the convention center (which is a great jogging path…) It’s a great place to stop and take a breath.
Beatles and Herring.
Chaos and Substance.
I also love ISC. I go to many of the events in the technical computing industry and ISC is a really interesting show. Each year great new ideas attack the dogma of HPC at ISC. “Commercial Off the Shelf will never make good HPC,” “Linux isn’t reliable enough,” “Nobody outside of the ivory towers needs HPC,” “Moore’s Law will run out before we get to Petascale,” and thud! Another bit of doctrine falls.
Each year I make my list of what I am looking for at ISC. What are the questions that need answering? Who is going to step up to create the buzz of the show floor? What new technology or idea will set aside the previously unassailable?
Here is my list for 2011:
- It’s the Workload Stupid: While the industry waits for major new platforms based on silicon coming in the second half of this year, there is still a great deal of innovation going on based on existing architectures. One of my pet theories is that there are many workloads in HPC that require distinct design points in hardware and software to achieve efficient and cost effective performance. Since there isn’t one form of parallelism or a canonical data structure, one size cannot fit all… Hence there is a thirst for architectural reinvention to deliver optimized performance. How are architectural innovators performing and how is their innovation being accepted to serve the distinct classes of workloads? Are major OEMs continuing to deliver new design points that target distinct HPC workloads?
- Alternate Architecture Acceptance: How have attached coprocessors like Intel’s MIC products and Nvidia and ATI GPGPUs been accepted as tools for delivering performance leading up to an exascale era? Are the advancements in software development methodology keeping pace with the complexity requirements and putting us on the glide slope we need as an industry? Intel has been signaling that we will see new proof points on its MIC architecture. Will Intel make its case that this architecture has legs? Will it make the case that the software model is viable?
Will HPC get its heads into the clouds? I have heard enough talk about HPC in the CLOUD. I want to see the industry start to pull together solutions. There is a relationship between HPC in the Cloud and the so called “Missing middle.” The missing middle is a moniker that refers to the many potential HPC customers who need HPC for technical computing, but cannot access it for technical, economic or social reasons. HPC in the Cloud could be one means of servicing these requirements. How will the industry bridge this gap? Intel has ideas here – But so do many others, watch this space.- Is FABRIC ripping at the seams: I am concerned that the rate of fabric development is not going to keep up with node level performance. Fabric remains a key bottleneck to HPC performance. What technologies are going to change the game in interconnect?
- Is Efficiency the Hobgoblin? Will the top500 list show any improvement in efficiency? Some of the systems to enter the top 10 have been hybrid systems with great peak flops, but pretty miserable efficiency numbers. Is this a trend that cannot be stopped?
Anyway, between the Beatles statues on the Raiperbahn and the windmills on the shore, A lot goes on in Hamburg. I think perhaps the industry will catch its breath and assess its challenges and opportunities, I hope it does… The path to exascale will consume us soon enough; we need the stamina to crash through.
IBM to Showcase Watson and Blue Gene at ISC’11
According to IBM’s latest newsletter, attendees at this year’s annual International Supercomputing Conference (ISC) will have the opportunity to learn how the company is delivering on their theme of High Performance Computing for a Smarter Planet.
As part of the exhibition, IBM will demonstrate their System x, POWER, and IBM Blue Gene technologies in addition to Watson, the supercomputing system that went from a modest DeepQA machine to a formidable contestant on Jeopardy! For Watson to rival the speed of its human competitors in delivering a precise answer to any given question required custom algorithms, terabytes of storage managed by GPFS, and thousands of IBM POWER7 computing cores working in a massively parallel system. Attendees will learn how this leap forward in data analytics technology is poised to impact business and science.
I’ve been working on my FlexRex parody of Watson for a while now, but it will be very cool to see this technology in person in Hamburg.
Interview: Eric Barton on Lustre Post-LUG and Plans for ISC’11
The recent LUG’2011 user group meeting for the Lustre community set the stage for the popular open source file system to move forward on the road to Exascale. To get an update on progress since then and preview of Lustre sessions at ISC’11, I caught up with Whamcloud’s CTO Eric Barton.
insideHPC: What’s happening with Lustre on the tech side at the moment?
Eric Barton: Performance is the key. Whamcloud is working on 2.x performance to match or exceed 1.8x. Lustre 1.8.x is extremely stable and well-proven in production. Lustre 2.0 involved a major re-structuring in preparation for future features and the focus was on stability. With Lustre 2.1 we are building on that stability and ensuring that the performance meets or exceeds Lustre 1.8.x. Other enhancements include RHEL6 server and client support, Async journal commits by default, and Ext4 by default. We also feel an adopted proven management model will enhance the workability and administration of a Lustre environment.
This is just some of the things we are working on. More updates to the roadmap and development are coming. For more details: http://wiki.whamcloud.com/
insideHPC: What are your plans for Lustre content at ISC?
Eric Barton: Whamcloud will have multiple technical experts at ISC to share updates on Lustre, the stability of the file system, and new developments, as well as how Whamcloud is contributing to the future of Lustre.
We have three presentations and chances to talk with the experts.
- On Sunday, June 19, 1:00pm – 1:30pm: Eric Barton, CTO, will be giving a talk at the HPC Advisory Council European Workshop entitled “Lustre Status and Roadmap.” For more information, check out the Workshop site. You’ll see a number of other great talks listed there as well.
- On Tuesday June 21, 2011, 12:40pm – 1:00pm: Brent Gorda, CEO, Eric Barton, CTO, and Johann Lombardi, Principal Engineer, will talk about the Lustre high-performance parallel file system and the specific work Whamcloud is doing to contribute to the longevity and stability of this core high performance computing technology.
- On Wednesday June 22: 02:30pm – 04:45pm Eric Barton, CTO, and Jacques-Charles Lafoucrière, EOFS, will lead a Birds of a Feather (BoF) discussion on “Lustre for HPC from Petascale to Exascale” including features under development, recent accomplishments and a look into the future roadmap of Lustre.
- And, last but not least, Whamcloud technology experts will be staffing the Whamcloud booth (#843) during all show floor hours at ISC.
insideHPC: Can you fill us in on the progress towards the next Lustre release?
Eric Barton: Whamcloud is leading the Lustre 2.1 community release, and we expect it to GA this summer. All up-to-date details can be followed at: http://groups.google.com/group/lustre-21
Although all future development effort is focused on the 2.x codeline, 1.8.x is still our maintenance release stream and the codeline most of the Lustre community uses in production. Therefore Whamcloud is also in the process of releasing a 1.8.x release that will support 32TB logical unit numbers (LUNs). GA is targeted for June 30th.
insideHPC: At LUG’2011, the main Lustre user communities agreed to align under OpenSFS pending changes to their bylaws. How is that progressing since LUG?
Eric Barton: At the Lustre User Group (LUG) on April 13 this year, LUG Chair Galen Shipman announced an agreement by the three Lustre community organizations to align with OpenSFS. We are extremely supportive of this move and believe that it is great news for Lustre to have all the community groups pulling in the same direction. Since things are changing quickly, for the latest progress and news, it’s probably best to use the http://www.opensfs.org site. Lots of news, information, and mailing lists are available there for anyone who is interested.
insideHPC: What keeps you up at night?
Eric Barton: Concalls with our global team 
Seriously, the big thing is the health and continued collaboration within the Lustre community. There were some tense moments at LUG 2011 where many were wondering what would happen with the main communities, specifically HPCFS, OpenSFS, and EOFS. We came together at LUG, things are good, and I have no reason to think it won’t continue along this positive path. But, I still worry.
Early Bird Registration for ISC’11 Ends May 20
The ISC early bird registration will close by the end of today, so be sure to register to save 25 percent off the price of admission. ISC’11 is looking to be another record year, with 22 percent more attendees registered compared with the same time last year.
The International Supercomputing Conference will take place in Hamburg from June 19 – 23. We’ll see you there!
Register Early for Savings – ISC’11 Tutorials and Workshops
The 26th International Supercomputing Conference is offering 11 educational tutorials and two brand new workshops – HPC in Asia and HPC Mid Market on Sunday, June 19 as a prelude to the main conference.
The five-day conference and trade fair will be held June 19 – 23, 2011, at the Congress Center Hamburg, Germany, bringing together around 2,000 IT managers, researchers and scientists to exchange knowledge and ideas, foster contacts and do business. The complete ISC’11 program comprises educational tutorial sessions, workshops, a conference and an exhibition consisting of 150 industrial and research exhibits. All conference proceedings are conducted in English.
Sunday Tutorials
The conference will kick off on Sunday, June 19, with three full-day and eight half-day tutorials on a broad range of application topics. The tutorials are led by expert speakers with hands-on experience in HPC, and the topics cover diverse areas of interest, e.g., architectures, algorithms and analysis, large-scale simulations, storage and data, as well as developing areas like Exascale computing, Clouds and CUDA. All 11 tutorials will be held in parallel sessions and the registration fee allows participants to visit all the sessions. More information is available here.
HPC in Asia Workshop
Last November, Asia astonished the global HPC community by deploying three of the five fastest HPC installations worldwide. To give ISC’11 attendees insight into these growing markets, ISC has organized a full-day HPC in Asia Workshop and invited experts from the most well-known HPC centers in China, Japan, Korea, Taiwan and Singapore to talk about the state-of-the-art developments in their regions. Among others, they will discuss current activities and projects in Asia in the “HPC Applications in Asia” and “Ultra Large Scale Systems in Asia” sessions. The day will end with a science tour of the German Climate Center (DKRZ) for all participants. Click here to view full program.
Mid Market HPC Workshop
Also held on Sunday June 19, the workshop on Mid Market HPC will provide specialized users with an overview of the available HPC solutions for small- and medium-sized enterprises, and also facilitate a dialogue between users, technology companies, hardware vendors, ISV’s and service providers.
Early Registration Ends May 20
Attendees who register by May 20 can take advantage of discounted early bird rates, saving up to 25 percent off the conference and exhibition fees. Attendees can tailor their registration using ISC’11’s flexible pricing system, selecting specific days or the full program to match their interests and schedule. Catering and evening events are included in the registration fees. Click here for information on registration.
This year, ISC will host over 150 industrial and research exhibitors from around the world between June 20-22, with organizations from Brazil, South Africa, China, Japan, Russia, the U.S. and most European countries participating. Read the exhibition overview.
Slidecast: ISC to Launch HPC in Asia Workshop, Spotlight on Exascale
In this slidecast, Rich Brueckner (insideHPC) and Mike Bernhardt (The Exascale Report) interview Martin Meuer, Director of Operations for ISC.
This year, the International Supercomputing Conference (June 19-23 in Hamburg) is launching and all-new HPC in Asia Workshop on June 19, 2011. Meuer describes the goals for the workshop along with plans for an exciting set of Exascale sessions at ISC.
At insideHPC, we are proud to be Gold Media Sponsors of ISC’11, so be sure to tune in here for special coverage of the event starting in June.
Download the Slides (PDF) * Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players
ISC’11 Keynotes to Explore the Leading Edge of HPC
ISC’11 in June will feature four compelling keynote presentations focusing on an array of HPC topics:
- Simulating the Brain – The Next Decisive Years, Henry Markram, EPFL Switzerland
- The Road to Exascale Computing – A European Challenge? Philippe Vannier, Bull, France
- HPC Achievement & Impact 2011, Thomas Sterling, Louisiana State University, USA
- Future Trends in Memory Systems: Showstopper or Performance Potential for HPC? Dean Klein, Micron Technology, USA
ISC will take place in Hamburg, Germany, from June 19 – 23. This year’s conference program looks terrific and I encourage you to check it out. Read the Full Story.
ISC’11 Issues Call for Tutorials
ISC’11 is gearing up for another great event this year in Hamburg, Germany June 19-23, 2011. Now the organizers have issued a Call for Tutorials. Specifically, they are seeking half-day or full-day tutorials with broad applicability.
The following topics are of specific interest to ISC:
- Parallelization for Heterogeneous Architectures including GPUs
- Hybrid Programming
- Parallel Applications/Parallel-I/O
- Parallel Programming Languages
- Visualization and Analysis for Very Large Data Sets
- InfiniBand/High-Speed Ethernet
- Performance Measurements and Optimization
- HPC and Clouds
Proposals are due by Tuesday, Feb. 15.
Call for Nominations: IDC’s HPC Innovation Excellence Award
Earl Joseph and the crew from IDC have issued a call for nominations for the first annual HPC Innovation Excellence Award Program. Designed to increase HPC adoption through the publication of ROI success stories, the program will recognize HPC users and their vendors for major achievements in industry, government and academia.
While there are multiple benchmarks to measure the performance of technical computers, there currently isn’t an adequate methodology to evaluate the economic and scientific value HPC systems contribute. The HPC Innovation Excellence Award Program is designed to help close that gap,” said Earl Joseph of IDC. “The award program provides an important means for gathering more quantified ROI success stories. Through identifying clear ROI examples, we expect to demonstrate more convincingly the close link between HPC and economic/scientific innovation and competitiveness.”
IDC hopes that the program will shed more light on new and emerging markets for HPC, such as the Missing Middle of small and medium businesses who could benefit from high performance computing, but haven’t made the plunge.
Attention vendors: Be sure nominate your customers by March 31, 2011. The first round of awards will be given out at the annual ISC conference the week of June 19, 2011 in Hamburg, Germany.
Download the nomination form in .docx format or PDF.
Last Chance: ISC’11 Call for Papers
The ISC’11 conference (June 19 – 23) is seeking papers reporting original work in theoretical, experimental and industrial research and development in the following areas: Architectures, Algorithms and Analysis, Future Trends, Large Scale Simulations, Storage and Data.
Hurry! Submissions are due January 23, 2011.
