Feeds

Americans invade Euro cluster compo

Buffalo, Bear-O-Dactyl vie for cluster glory

Choosing a cloud hosting partner with confidence

ISC 2012 Two American universities vie for cluster fame and glory in the inaugural ISC'12 Student Cluster Competition (iSCC) in Hamburg next month. This competition, sponsored by Airbus, the HPC Advisory Council and ISC, gives university teams the chance to design, build, and benchmark their clusters against four teams of peers. They’re limited only by a hard power cap of 13 amps and the requirement that the gear they use be commercially available.

(For more on the competition in general, click your clicker here.)

The University of Colorado team has more institutional experience in the student clustering game than all of the other competitors combined. They’ve competed in all five of the SC student cluster challenges, but have yet to win all the marbles (there aren’t any real marbles), although they did score as ‘Fan Favorite’ in New Orleans at SC10.

Last year in Seattle, they brought more memory per node (126GB vs 64GB), a huge number of AMD Interlagos cores, and also the tallest team member (at 7ft) in competition history. Despite also having a veteran-packed team and coaching staff, solid vendor support, and Buffalo spirit (not sure what this is, but it seems to fit) they didn’t fare better than a middle-of-the-pack finish.

The Buffs have been great competitors. They show up on time and are unfailingly polite – they’re always nice – maybe too nice to win a competition like this? Do they have the killer instinct and drive necessary to conquer their cluster foes? Will we see a different Team Colorado in Hamburg next month?

Stony Brook University, the second US-based team, has been to the big cluster dance before. They were there for the first SC competition way back in '07, and won the overall crown (there isn’t a real crown) at SC09 in Portland.

They returned to New Orleans in 2010 to defend their title with a Cray-based design that was one of the first systems to utilise GPUs. They also dumped their university mascot (Wolfie the Seawolf) in favor of their own creation, the Stony Brook Clustering “Bear-O-Dactyl”. (“Why a ‘Bear-O-Dactyl?” “Because it’s scarier than a bear or a pterodactyl on their own.” Good call.)

Unfortunately, the combined power of the GPU and the Bear-O-Dactyl, supplanted by some mil-spec caffeine brought by one team member (yep, saw the box myself, and it was pretty cool) wasn’t quite enough to get them over the hump. They were beaten by Taiwan for the overall performance title, topped by Texas in LINPACK, and lost the Fan Favorite award to Colorado.

The members of Team Stony Brook were quirky and willing to take chances to leapfrog the competition. Their gamble with GPUs didn’t pay off in New Orleans; they just didn’t have enough optimised code, and the rest of their system wasn’t strong enough to make up for the performance shortfall. Will Hamburg be a different story? Can Stony Brook regain their former glory?

The Colorado Buffaloes and Stony Brook Seawolves have their work cut out for them in Hamburg. While both teams have had some competition experience, they’ll be facing a pair of formidable Chinese teams and the German home team. This will also will be the first time they’ve played outside the US.

Will the long trip have an impact on their efforts? The beer in Germany is almost certainly heavier than they’ve experienced in the US; will that play a role? Germany, like the rest of Europe, puts a near-fatal amount of power through wall outlets (220 or 240 volts vs 110 or 120 in the US). Will this become a factor in the competition?

According to research that I can’t find right now, as many as 47 per cent of American tourists in Europe are either killed or severely injured by electrical mishaps associated with plugging/unplugging common appliances. This hasn’t received much media attention, probably due to lobbying efforts from the duty free stores and the GSTC (Global Souvenir & Trinket Consortium). ®

Secure remote control for conventional and virtual desktops

More from The Register

next story
Azure TITSUP caused by INFINITE LOOP
Fat fingered geo-block kept Aussies in the dark
NASA launches new climate model at SC14
75 days of supercomputing later ...
Yahoo! blames! MONSTER! email! OUTAGE! on! CUT! CABLE! bungle!
Weekend woe for BT as telco struggles to restore service
Cloud unicorns are extinct so DiData cloud mess was YOUR fault
Applications need to be built to handle TITSUP incidents
NSA SOURCE CODE LEAK: Information slurp tools to appear online
Now you can run your own intelligence agency
BOFH: WHERE did this 'fax-enabled' printer UPGRADE come from?
Don't worry about that cable, it's part of the config
Stop the IoT revolution! We need to figure out packet sizes first
Researchers test 802.15.4 and find we know nuh-think! about large scale sensor network ops
DEATH by COMMENTS: WordPress XSS vuln is BIGGEST for YEARS
Trio of XSS turns attackers into admins
SanDisk vows: We'll have a 16TB SSD WHOPPER by 2016
Flash WORM has a serious use for archived photos and videos
prev story

Whitepapers

Why cloud backup?
Combining the latest advancements in disk-based backup with secure, integrated, cloud technologies offer organizations fast and assured recovery of their critical enterprise data.
Forging a new future with identity relationship management
Learn about ForgeRock's next generation IRM platform and how it is designed to empower CEOS's and enterprises to engage with consumers.
Designing and building an open ITOA architecture
Learn about a new IT data taxonomy defined by the four data sources of IT visibility: wire, machine, agent, and synthetic data sets.
How to determine if cloud backup is right for your servers
Two key factors, technical feasibility and TCO economics, that backup and IT operations managers should consider when assessing cloud backup.
Reg Reader Research: SaaS based Email and Office Productivity Tools
Read this Reg reader report which provides advice and guidance for SMBs towards the use of SaaS based email and Office productivity tools.