October 31, 2013
Oct. 31 -- Scientists at The University of Manchester have used Polaris, the N8 Research Partnership High Performance Computing (N8 HPC) facility, to simulate for the first time ever how one of the world’s largest dinosaurs would have walked.
The analysis has revealed that Argentinosaurus, which weighed more than 80 tonnes, would have walked with a slow steady gait. The research, published in PLOS ONE, is important for understanding more about musculoskeletal systems since all vertebrates, from dinosaurs to humans to fish, all share the same basic muscles, bones and joints.
N8 HPC, which is funded by the Engineering and Physical Science Research Council (EPSRC), provides a high performance computing facility service for the Universities of Durham, Lancaster, Leeds, Liverpool, Manchester, Newcastle, Sheffield and York – an established collaboration known as the N8 Research Partnership – and industry partners. Capable of a peak performance of 110 trillion operations per second – the approximate equivalent to half a million iPads - it enables academic and private sector researchers to build more realistic models involving large amounts of data and to undertake more complex analyses in many research fields, including life sciences, energy, digital media and aerospace.
Dr. Bill Sellers, lead researcher on the project from the University of Manchester, Faculty of Life Sciences said, “If you want to work out how dinosaurs walked, the best approach is computer simulation. This is the only way of bringing together all the different strands of information we have on this dinosaur so that we can reconstruct how it once moved.
“To understand how muscles, bones and joints function, we can compare how they are used in different animals. Argentinosaurus is the biggest animal that ever walked on the surface of the earth and understanding how it did this will tell us a lot about the maximum performance of the vertebrate musculoskeletal system. We need to know more about this to understand how it functions in ourselves.
“Similarly, if we want to build better legged robots then we need to know more about the mechanics of legs in a whole range of animals, and nothing has bigger, more powerful legs than Argentinosaurus.”
The £3.25m N8 HPC facility is a Tier 2 SGI 5,000+ core high performance computing cluster, with 332 compute nodes. Each node has two of the latest generation Intel E5-2670 ‘Sandy Bridge’ processors and these nodes have a capacity of 320 GigaFLOPS. By using a Mellanox QDR InfiniBand interconnect to join all of Polaris’ nodes together, a peak performance of 110 TeraFLOPS is possible, making it one of the 250 most powerful computers in the world.
Dr. Lee Margetts, of The University of Manchester Research Computing Services, said, "Access to the N8 HPC system was a critical factor that enabled the team to finish the research in time for the PLOS ONE Special Collection on Sauropods. Timing is important as this collection is likely to be the "de facto" international reference for Sauropods for decades to come. The researchers report that they were very impressed by the system as their software ran twice as fast on N8HPC than HECToR, when using the same number of cores."
The team of scientists also included Dr Rodolfo Coria from Carmen Funes Museum, Plaza Huincal, Argentina, who was behind the first physical reconstruction of this dinosaur that takes its name from the country where it was found. Dr Phil Manning, from Manchester who also contributed to the paper, said: “It is frustrating there was so little of the original dinosaur fossilized, making any reconstruction difficult. The digitization of such vast dinosaur skeletons using laser scanners brings Walking with Dinosaurs to life…this is science not just animation.”
The University Manchester team now plans to use the same method to recreate the steps of other dinosaurs including Triceratops, Brachiosaurus and T.Rex.
Source: University of Manchester
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?