Allinea
Xyratex International
HPCwire

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Language Flags

People to Watch 2013
Datanami
Digital Manufacturing Report
HPC in the Cloud
Green Computing Report

Tabor Communications
Corporate Video

Blog: From the Editor

From the Editor | Main Blog Index

The Essential ISC


As I get ready to head to Hamburg for the 24th running of the International Supercomputing Conference, I'm taking a last look at the conference agenda to make sure I don't miss any interesting bits. Fortunately the conference maintains a plenary track of sessions, but there are other activities -- BoF sessions, exhibitions, poster sessions, and the obligatory networking and meetings -- taking place in parallel, so some planning is mandatory. With that in mind, I thought I'd point to the some of the highlights of the four-day supercomputing extravaganza.

Of course the opening session on Tuesday includes the much-anticipated announcement of the latest TOP500 list. I've got to admit, I love the horsepower race, but this year might not have the excitement of recent lists, when petaflop supers were breaking onto the scene. Unless IBM manages some last-minute surprise, ORNL's Jaguar super will take over the number one spot from the Roadrunner machine at Los Alamos. There is also the expectation that the rest of the top 10 will be shuffled as the newest international supers find their place.

The most interesting aspect to watch may be the TOP500 list's turnover rate, which has accelerated recently. Because of the crummy economy though, some are predicting a dip in TOP500 volatility. The turnover, or lack thereof, is not likely to excite the mainstream IT press very much, but this is something I'll be watching.

At the end of the opening session is Andy Bechtolsheim's keynote, which focuses on the evolution of HPC interconnects. Bechtolsheim, who co-founded Sun Microsystems and is now the chief development officer at Arista Networks, seems to have impeccable timing with industry trends. His original $100,000 investment in Google is now thought to be worth over $1 billion. In 2007 he co-founded 10GbE switch maker Arista Networks, just as the fortunes of his original firm, Sun Microsystems, seemed to be heading south. Bechtolsheim is always worth listening to.

Later on opening day, you have a choice of sessions: one on HPC climate modeling and the other on supercomputing challenges in aeronautics, as well as a gaggle of scientific research sessions and BoF meetings. Unless you have a particular research interest in aeronautics or one of the super-science topics, I'd go for the climate modeling session, and in particular, the panel discussion in the afternoon: "If Global Change is the Grand Challenge Application, Do We Need a World Climate Computing Center?" Great question, and in the age of supercomputing one-upmanship, it implies we'll need some political finesse to build a global HPC capability.

Finally, don't miss the opening day welcome party, starting at 6:00 p.m. and presumably ending when the food and beer run out. Good chance to chat with the exhibitors, future employers, close friends and despised enemies. It's not as exhausting as it sounds. In many cases, these are all the same people.

Which brings us to Wednesday. If you only go to one presentation the whole week, I'd recommend the cloud computing session Cloud Computing in HPC -- Synergy or Competition. Current deployments and use cases of cloud technology, and assessments of how the cloud model fits with high performance computing will be discussed. There's an all-star cast of HPC movers and shakers coming from traditional vendors (IBM, Sun, HP and Microsoft) as well as the high-flying cloud companies (Google, Amazon and Yahoo). A panel discussion at the end wraps it up. Two of the panel participants are Google and Microsoft, so there is a possibility of gun-play.

Later that day is the first hot seat session, which puts some of the top brass at 18 HPC companies on the spot. The vendors are given 10 minutes to speak their piece and then field a couple of questions from "hand-picked inquisitors," most of whom come from the HPC research community. On Wednesday, Microsoft, Intel, IBM, Bull, Spectra Logic, Fujitsu, BlueArc, Cluster Resources, and Cray get grilled. On Thursday, the second hot seat session puts Sun Microsystems, Supermicro, HP, SGI, Mellanox, Myricom, QLogic, Force10, and NEC under the spotlight. An ISC tradition.

Also on Wednesday, LSU professor and HPC maven Thomas Sterling will deliver his customary state-of-the-supercomputer address for 2009. Sterling is always informative and entertaining. Not to be missed.

On Thursday, a couple of academicians try to answer the existentialist question for the multicore obsessed: What the heck is with all these cores? The multicore session features Yale N. Patt, Professor of Electrical & Computer Engineering at the University of Texas at Austin, and Kathy Yelick, Director of NERSC and CS prof at UC Berkeley. Patt will talk about future architectural directions of multicore processors, while Yelick will give us the latest developments on the software side.

For something completely different, you'll want to check out IBM's Prof. Dr. Gunter Dueck's keynote on Thursday afternoon. His presentation is about Lean Brain Management. What pray tell is that, you ask? From the session summary:

Lean Brain Management aims at economizing on intelligence by moving all the necessary brain work in to a perfect system which can be handled by completely unskilled workers, the future "morons". Only the system must be intelligent, not the employees. We estimate that a perfect system can be designed by a few intelligent people, the so-called "moronorgs", who dumb down all business processes to perfect foolproof Lean Brain Quality.

Groovy. Presumably mind-expanding drugs will be handed out before the talk.

On Friday, make sure you attend the morning session on what's being done with the new petaflop (and a few near-petaflop) machines around the world. In this session, we hear from some of the masters of the petaverse from the US, China, India, Germany, and Switzerland.

That wraps up my hot picks. To see the entire conference schedule, take a look here.

Posted by Michael Feldman - June 18, 2009 @ 3:53 PM, Pacific Daylight Time

Michael Feldman

Michael Feldman

Michael Feldman is the editor of HPCwire.

More Michael Feldman

Mellanox

Recent Comments

No Recent Blog Comments

Feature Articles

Jordan Unveils PS3-based Supercomputer

The Jordanian government has requisitioned a custom-built supercomputer, which is now being used by universities across the country. What it's made of may be surprising – but it's not unprecedented.
Read more...

The Week in HPC Research

The top research stories of the week include novel methods of data race detection; a comparison of predictive laws; a review of FPGA's promise; GPU virtualization using PCI Direct pass-through; and an analysis of the Amazon Web Services High-IO platform.
Read more...

NetApp, SGI Flex HPC Storage Muscle

NetApp and SGI made simultaneous announcements yesterday of new storage systems specifically targeted at HPC and big data applications. NetApp launched its latest offering in the E-Series line of controllers, the NetApp E5500, and OEM partner SGI adopted the technology for its InfiniteStorage 5600 RAID system.
Read more...

Short Takes

Intel Xeon Phi Versus 'Sandy Bridge'

Mar 05, 2013 | How does the Phi coprocessor measure up to Xeon "Sandy Bridge" brand-mate?
Read more...

UberCloud HPC Experiment Preps for Round Three

Mar 05, 2013 | The cloud-based experiment seeks participants from the HPC, CAE, life sciences, and big data communities.
Read more...

StarCluster 101

Mar 01, 2013 | Create DIY HPC clusters in the Amazon cloud with the StarCluster toolkit. Here's how.
Read more...

World's Fastest Supercomputer Hits Speed Bump

Feb 27, 2013 | When it comes to Titan's final acceptance testing, ONRL says not so fast.
Read more...

How to Explain Your Job to Your Kids

Feb 27, 2013 | On the lighter side of HPC, here's a video explaining what supercomputers do.
Read more...

Sponsored Whitepapers

High-Performance Computing in Action

03/07/2013 | HP and Intel | Businesses that want to be on the cutting edge of their industries are increasingly turning to high-performance computing (HPC) solutions to handle complex compute processes and speed up their rate of innovation. Download this Executive Brief to see how businesses in energy, life sciences and entertainment put HPC solutions to work in their operations.

Case Study: Allinea Software Saves Climate Change Researchers Precious Time

02/11/2013 | Allinea | Researchers have no time to lose in their quest to divine the future of the Earth’s climate and its effects on humanity. When a problem arises, one of the world’s top centers for climate research relies on Allinea DDT.

Sponsored Multimedia

Webinar: Extreme Scalability for High-Fidelity CFD Simulations

CFD models of 100 million+ cells are increasingly common. Models of a billion+ cells are emerging. To meet demand for 10 to 100x larger models, the simulations will have to scale to 10 to 100x more cores. Both compute environment capabilities and capacity must be enhanced. Learn how Cray and ANSYS are scaling production models to 1000s of cores and providing the compute environment to take CFD simulations to the next level of extreme fidelity.

Webinar: Methodology & Tools for Compute Performance at Any Scale

Clusters and workstations are often sized by one-off benchmarking campaigns during tenders, way before production. This method is efficient, but has drawbacks. This webinar will expose HP’s benchmarking method and the principle and benefits of leveraging this method day-to-day production. It will highlight some unique intellectual property, namely HP iLO 4 and HP Cluster Management Utility (CMU), to perform “smart job scheduling”. Register Now.

Blogs by Topics

Blogs by Author

HPC Blogroll

Aspen

Featured Events






  • June 16, 2013 - June 20, 2013
    ISC'13
    Leipzig,
    Germany


HPC Wire Events