HPC > More stories

iVEC's Magnus supercomputer

iVEC seeks testers for petascale powerhouse

As it prepares to ramp up to petascale level, the iVEC supercomputer facility in Western Australia is looking for early adopters to run its iron through its paces. iVEC's current Magnus machine, a Cray XC30 with 104 blades (four nodes per blade, two 8-core Intel Sandy Bridge processors) comprises a total of 3,328 cores …

Find a partner? Nah, these students ALL play with same tools in giant CLUSTER-SHUCK

HPC blog One of the major differences between the recently completed Asia Student Supercomputer Challenge and the other major international competitions is the way the students get their clustering gear. In the International Supercomputing Conference (ISC) and the SC competitions, part of the task for students is to find a vendor partner …

Kiwis unplug supercomputer after intrusion

New Zealand's National Institute of Water and Atmospheric Research (NIWA) has deployed perhaps the ultimate security device - an RJ 45 plug's locking tab - to protect its 34 Tflops FitzRoy supercomputer after its security was breached. The locking tab was employed after the $NZ12.7 million machine, installed in 2011, was …
The Large Hadron Collider on Google StreetView

CERN: Build terabit networks or the Higgs gets it!

CERN's openlab has published a document explaining stuff it wants someone to invent so it can get the most out of the large hadron collider (LHC) and its other instruments. The Whitepaper on Future IT Challenges in Scientific Research (PDF) ) sets out the problem, namely that exotic kit was needed to cope with the petabyte-per- …
Simon Sharwood, 23 May 2014
CatalystNVRAM

Livermore lab opens Catalyst super to industry users

The Lawrence Livermore National Laboratory “Catalyst” supercomputer, which started sucking its first electrons in November 2013, is now open for industry workloads. When Igor pulled the big red switch last year, the lab touted features like the 800 GB of Flash attached to each of its 304 nodes via PCIe, in addition to the per- …
Nvidia

Nvidia upgrades PGI 2014 accelerator compile suite

Just a few months after first releasing its PGI 2014 set of compilers and tools, Nvidia subsidiary The Portland Group (PGI) has expanded OpenACC C++ support in the high-performance computing compile suite, and put older versions of Windows on the EOL list. The PGI 2014 release back in February added support for the latest …
iVEC's petaflops-housing Pawsey Centre data center

Australia to build 35,000-core supercomputer on Xeon-E5-2600 v3

Australia's Pawsey Supercomputing Centre has revealed basic information about an upgrade to its 'Magnus' supercomputer. The upgrade will see Magnus transmogrify into a Cray XC30 supercomputer with over 35,000 cores. Intel's future Intel Xeon processor E5-2600 v3 will be pressed into service to give Magnus its upgrade and take it …
Simon Sharwood, 11 May 2014
Galaxy simulation

ENTIRE UNIVERSE created in supercomputer. Not THIS universe (probably)

Researchers at MIT and Harvard have created a a giant virtual universe – 350 million light-years squared – which can be used to simulate the real universe's development from 12 million years after the Big Bang to the present day, or around 13 billion years' worth of expansion. Dubbed Illustris, the pocket universe uses over 12 …
Iain Thomson, 8 May 2014

Experienced HPC compo kids welcome eager Brazilians to cluster-wrestling match

HPC blog In our last story, we looked at the local Chinese undergrads who are competing for cluster-building glory at the Asian Student Supercomputer Challenge (ASC14) in Guangzhou, but there are plenty of competitors who are keen to grab the title and take it home – to the United States, steamy Brazil or even to the slopes of the Urals …
Stanford's Neurogrid board

Boffins build billion-synapse, three-watt 'brain'

A top-notch supercomputer can beat humans at many things, but they're also energy hogs – which is one reason so much work goes into neural simulation as the basis for computers. Now, Stanford scientists have demonstrated a neural simulation which they say has a million neurons, multi-billion synapse connections – and runs on …

Junior gurus go gung-ho in Guangzhou: Meet the students putting a boot in clusters' thrusters

HPC Blog That question everyone is asking: who are the undergrads competing for cluster-building glory at the Asian Student Supercomputer Challenge (ASC14) in Guangzhou this week? It’s a mixed slate of former cluster competition champions, wannabes, and newbies from almost every continent on Earth. First, make sure you're up to speed on …

Sixteen hot students, 3 kilowatts, $29k in CASH: It's cluster combat, China style

HPC Blog We're in Guangzhou, China, for the final round of the largest student cluster competition in the world. And to make it even more tense, there's nearly $28k up for grabs... The third annual Asian Student Supercomputer Challenge (ASC14) features sixteen university-backed teams from around the globe, competing to see which team …

Big Blue's GPFS: The tech's fantastic. Shame about the product

Storagebod IBM is a great technology company: so many of the technologies we take for granted can be traced to back to Big Blue. And many of today’s implementations still are poorer than the original implementations. And yet the firm is not the dominant force it once was; an organisational behemoth, riven with politics and fiefdoms doesn’t …
StorageBod, 15 Apr 2014
China's Tianahe-1A Supercomputer

Russia, India aim to dislodge China from No 1 on top of the 'flops

Russia and India may be about to join forces in taking on a common adversary: China and its floptastic Tianhe-2 beast which currently sits atop the Top 500 supercomputer list. Boris Shabanov of the Russian Academy of Sciences (RAS) has invited a group from the Indian Institute of Science and local Karnataka lawmakers to discuss …
Fujitsu SoC

New Fujitsu SoC designs shrink supers' power slurp

Fujitsu is pimping system-on-chip (SoC) designs for two new supers it says get 40 per cent more processing punch while slurping half as much power. The company also says its GS21 2400 and GS21 2600 need 70 per cent less data centre space. The SoC devices at the heart of the two servers consolidate 14 chipsets, Fujitsu says. …
SGI supercomputer with liquid cooling

Eco-friendly fluid keeps SGI supercomputer cool and moist

Intel and SGI have built a new supercomputer using an advanced coolant from 3M flowing directly over the hardware to slash the running costs of such systems, and possibly to boost performance. SGI supercomputer with liquid cooling Don't try this with water The SGI ICE X system uses Intel Xeon E5-2600 processors for grunt, and …
Iain Thomson, 9 Apr 2014
T Platforms GPU Blade

Sod Bitcoin - solving graph problems is the new GPU-flogger

HPC blog These GTC conferences, the annual GPU-fest sponsored by NVIDIA, are my favorite events of the year. Why? It’s how the event is organized. At GTC, real-world researchers, analysts, and other customers present the vast majority of sessions, not corporate mouthpieces. They talk about their projects, the challenges they’ve dealt …

Need a refresher on virtualisation?

HPC Blog This year at GTC, NVIDIA was pushing its Kepler-based GRID GPU tech to virtualise desktops, enabling a “anything, anywhere, anytime, on any device” usage model, but as a side effect delivered lots of guidance for VDI and virtualisation newbies. GRID was one of the big announcements at GTC12, with demonstrations of the tech's …
WASP-12b getting devoured by its sun

Sun-seeking NASA boffins steer FLEETS of satellites... BY HAND

HPC blog Looking for something to test your Star Wars video game skills? How about this: launch four satellites and fly them in eccentric orbits around the Earth. Each satellite is 1km apart from its neighbour, which sounds like a lot of space until you consider that they’re travelling at several kilometres per second during the …

HPC boffins: Our HOT MODELS will help you cope with DISASTER

HPC Blog What’s the most common and costly type of natural disaster? If you answered "floods and tsunamis” you’d be right on the money. Economic damages caused by large and small flooding events – either caused by storms, storm surges or earthquakes, in the case of tsunamis – are truly massive, running into the billions or tens of …

Get ready for software-defined RADAR: Jam, eavesdrop, talk and target ... simultaneously

HPC blog With a big RF transmitter and enough fast computing power, you have the ability to do a lot of different things, as evidenced by a General Electric presentation on "software-defined radar" at the GPU conference this year. At GTC 13 last year, GE gave a standing-room-only presentation about how it's using RDMA (Remote Direct …
REM's Shiny, Happy People

VDI a 'delightful' experience... Really?

GPU Technology Conference The VDI talk was the kind of GTC session I love. It’s where a real-world expert talks about how a difficult task is actually accomplished. Not the theory, not how it should work on paper, but what it takes to actually move a project from Point “A” to Point “We’re done with this”. Ken Fingerlos from Lewan Technology delivered in …
Nvidia CEO Jen-Hsun Huang provides details on next-generation GPU architecture, 'Pascal'

Nvidia unveils Pascal, its next-gen GPU with hella-fast interconnects and 3D packaging

GTC Nvidia has announced its next-generation GPU, and with it a new interconnect it dubs NVLink that it says will increase the bandwidth among GPUs – and also CPUs, should a processor designer implement it – over that of PCIe by five to twelve times. Nvidia NVLink block diagram String a bunch of GPUs together with NVLink, and you …
Rik Myslewski, 25 Mar 2014
Charlie Rose (L) and Larry Page (R) at TED 2014 in Vancouver

Google's Larry Page: DeepMind 'discovered CATS on its own'

Google co-founder Larry Page has said he was "disappointed" with the US government after NSA whistleblower Edward Snowden revealed just how much the agency was snooping on everybody. Charlie Rose (L) and Larry Page (R) at TED 2014 in Vancouver Charlie Rose (L) and Larry Page (R) at TED 2014 in Vancouver. Credit: James Duncan …

SPEC puts forward benchmark for accelerated supers

SPEC – the Standard Performance Evaluation Corporation – is pimping a new set of benchmarks it says will help HPC users get the most out of their hardware-accelerated systems. The “new benchmark suite that measures the performance of systems using hardware accelerator devices and supporting software”, SPEC ACCEL 1.0, comprises …
The Zythos Supercomputer at Australia's iVEC

What could you do with 264 Sandy Bridge cores and 6TB of RAM?

What could you do with 264 Xeon E5-4610 cores running at 2,5GHz and 6TB of RAM? As it happens, you can now try to find out: Perth's supercomputer facility iVEC has just shouted-out to anyone who likes the idea of taking its new, large-memory computer system Zythos for a spin. Zythos is an SGI UV2000. The machine is built from …
Simon Sharwood, 16 Mar 2014
SKA antennas close-up - artist's impression

Square-kilometre radio telescope wins millions in UK funding

UK science minister David Willetts has announced £11m of funding to help the Square Kilometre Array (SKA) process the expected deluge of data it will produce. Handling the expected 915-petabyte daily data firehose from the entire SKA radio telescope will require advances in practically every computing discipline: the data will …
Samsung 500GB SpinPoint 2.5in HDD

Traditional RAID is outdated and dying on its feet

HPC blog The video below is an interview I did with IBM storage guru Robert Murphy at SC13 in Denver. I’m still in catch-up mode after my recent near-disastrous rootkit episode. Youtube Video In the video, Robert and I talk about how today’s typical RAID mechanisms (1, 5, 6, 10) just aren’t up to the job of protecting data against drive …
Cray Mongolian cluster

Supercomputer maker Cray: Jobs for HPC kids? There might be some going

HPC blog There are some teens and kids in their twenties who skip out on university courses for a trip abroad with their friends, where they spend days and nights chugging down (caffeine) brew, losing sleep and wrestling with, er, clusters. But this isn't necessarily a bad thing... In fact, it could mean that they'll one day work at Cray …

Reg HPC man relives 0-day rootkit GROUNDHOG DAY

HPC blog This is a difficult article to write, and I’ve put it off for way too long. But it’s time to bite the bullet and make an embarrassing admission to the Register audience. I’ve been hacked and hacked hard. Admitting this publicly to Reg readers is like chumming shark-infested waters with my own blood. Or like telling people that I …
Windows 9x BSOD

How NOT to evaluate hard disk reliability: Backblaze vs world+dog

HPC blog A few months ago, Brian Beach, a distinguished engineer at cloud backup joint Backblaze, published a set of study-like blog postings relating to his firm's experiences with hard drive lifespan in its 25,000+ spindle environment. The blogs garnered quite a bit of interest due to the subject matter, and provocative titles like: …
Shot of the D-Wave chip holder - a robotic looking contraption

D-Wave wooing universities down under

While controversy continues over the nature of its machines, quantum computing company D-Wave is wearing out the shoe-leather talking to academic users – and The Register. In Australia to present to universities and the HPC community at large about D-Wave, the company's director of business development and strategic partnerships …
IBM's Watson playing Jeopardy

IBM to invest $100m and a decade into using Watson in Africa

IBM is planning to invest $100m and 10 years on getting Africa hooked on its Watson supercomputer system. Big Blue said that it wants to use the system in "Project Lucy", named after the earliest known human ancestor, to help the continent in key areas like healthcare, education, sanitation and agriculture. "In the last decade …

Open MPI hits milestone with FORTRAN-ready 1.7.4 release

The decade-old Open MPI project, beloved of the HPC community, has shipped code for a major feature release that brings it close to a complete MPI 3 implementation. Speaking to The Register, Cisco's Jeff Squyres (who along with Richard Graham, then of Los Alamos and now with Mellanox, was one of the instigators of the Open MPI …
OCP Storage VAult

OCP gets Ethernet silicon interface

HPC interconnect specialist Mellanox has added an Ethernet switch SDK to its contribution to the Facebook-led Open Compute Project. The SDK follows contributions dating back to 2013 to the OCP's networking sub-project, an effort which also attracted participants like Broadcom, Cumulus Networks, and Intel. For that project, …
Shot of the D-Wave chip holder - a robotic looking contraption

Boffins say D-Wave machine could be a classic*

First, the world thought that D-Wave hadn't built a quantum computer; then, it thought there was a quantum computer in the box; next, there was disappointment that the D-Wave machine didn't speed things up (but might still be quantum); and now, it starts to look like it's not quantum after all. In the latest instalment in the …
cloud

HPCaaS experiment becomes commercial service

After 18 months of experimentation, UberCloud has launched a commercial service to help academic HPC users find cloud-accessible resources for their projects: a kind of HPC-as-a-service matchmaker. The project follows two lines of thought: although there's plenty of compute cycles already sitting on the cloud courtesy of Amazon …

My other supercomputer is a Lenovo: What IBM System x sale means for HPC

HPC blog Back in May 2013, when IBM was first rumoured to be selling part of its System x division to Lenovo, I wrote a story looking at how this might impact IBM’s position in the HPC market. It’s a good time to revisit that analysis and see what the picture looks like now. As I noted in the prior story, it’s damned hard to get solid …
Donkey Kong

King Kong's dangling cluster cut loose

The supercomputer used to animate Peter Jackson's 2005 version of King Kong has been donated to New Zealand's Auckland University of Technology (AUT). The supercomputer comprises 200 IBM HS20 Blade servers, a piece of kit Big Blue sold back in 2005 and a couple of preceding years. IBM's page for the HS20 suggests the machines …
Simon Sharwood, 24 Jan 2014
The ISS shown transiting the Sun. Pic: Thilo Kranz/DLR

'We don't use UPS. If we did we'd have huge UPSs and tiny computers'

The heatwave-driven outage at the VLSCI supercomputing facility last week could have been worse than it was, with power cuts also a risk, the facility has confirmed. A senior systems administrator at VLSCI, Chris Samuel, has discussed the outage and the lessons learned with The Register. While the reason for the shutdown was …
Shot of D-Wave's 16 qubit chip

Boffin benchmark battle after D-Wave quantum kit crawls in test

Yet another attempt to benchmark the D-Wave quantum computer – this time, in its 512 qubit configuration – has come up with disappointing results, with the company responding that it was the wrong kind of test. The work, led by Matthias Troyer of ETH Zurich, pitted a D-Wave Two (the machine that Google bought to much fanfare …
The Bureau of Meteorology's heatwave forecast

'Climate change' event dishes up sous vide supercomputers

Call it climate or call it unusual weather, but last week's five-day heat wave in the Australian city of Melbourne got the sweat running at one of the country's major supercomputing facilities, the Victorian Life Science Computation Initiative (VLSCI). The problem for sysadmins at VLSCI was simple: with temperatures remaining …