This site may earn affiliate commissions from the links on this page. Terms of use.

BlackHoles1

Hewlett Packard Enterprise has appear a new collaborative effort with the Stephen Hawking Centre for Theoretical Cosmology. The new effort, which leverages HP's Superdome Flex servers, is a step forrad for the Cosmos projection, which has existed in one form or another since 1997. Back then, SGI and Intel were both major players, though HPE seems to be tackling the upgrades and improvements on its ain so far.

"Our Cosmos group is working to understand how space and time work, from before the first trillion trillionth of a 2nd subsequently the Big Bang up to today," said Hawking, the Tsui Wong-Avery Managing director of Research in Cambridge'south Section of Applied Mathematics and Theoretical Physics, in a statement. "The recent discovery of gravitational waves offers astonishing insights near black holes and the whole Universe. With exciting new data similar this, nosotros need flexible and powerful estimator systems to keep ahead so we can examination our theories and explore new concepts in primal physics."

HPE says this new Superdome volition bring together an existing HPE Apollo supercomputer, just it'll scarcely be handling the heavy lifting all on its ain. Cosmos already has an SGI Altix UV2000, Cosmos2 (1536 Xeon E5-4650L), and Cosmic (288x Intel Xeon E5-4650L, forth with 24 Intel Xeon Phi MICs (5110P).

group_shot_large

The Centre for Theoretical Cosmology's staff.

HPE's Superdome Flex is unquestionably a beast; the machine can scale from 4-32 sockets and supports 768GB-48TB of memory. Exactly what benefits this will deliver to the Cosmos projection is unknown. While HPE makes much of its in-retention calculating (meaning information technology holds huge datasets entirely in DRAM), information technology'southward non clear how much of a specific benefit this provides. NextPlatform has done a good deep dive into in-memory computing, pointing out that while it's indisputably expert for solving certain kinds of bug, it's not the panacea HP markets it as being: "The glib way to say it is that in-retention does work if everything fits, and when it doesn't fit, it actually doesn't fit and you lot have got a problem," NextPlatform writes.

superdomex

Superdome Flex. Credit: HPE

DRAM also costs much more per byte than conventional storage and draws considerably more power than NAND flash or magnetic disks. It's as well not clear if the COSMOS project can brand utilize of the platform's in-memory computing if other systems on the same network aren't optimized to do so as well. Then again, one of the major differences betwixt an HPC cluster and a conventional platform is that it's often worth it to perform all-encompassing customization on a system for HPC workloads, thanks to the long-term improvements such optimization can deliver.

The new Superdome Flex system won't be limited to COSMOS, just will also exist available to numerous other research departments.

"Loftier performance computing has become the third pillar of research and nosotros wait forward to new developments across the mathematical sciences in areas as diverse as sea modeling, medical imaging and the physics of soft matter," said Professor Nigel Peake, Head of the Cambridge Department of Applied Mathematics and Theoretical Physics.

Pinnacle image: NASA/JPL-Caltech