insideHPC » New Installations

Video: Open Science Rides LANL Mustang Super

Rich Brueckner — Thu, 09 Feb 2012 08:02:38 +0000

www.youtube.com/watch?v=VLuYuBqXdWY

In this video, Bob Tomlinson from LANL describes the recently acquired Mustang supercomputer from Appro.

The Mustang system has been supporting larger jobs as was intended and in just few months of use it has already offered 434 Million CPU hours for competitive, peer-reviewed, open science. According to the latest November 2011 Top500 list, Mustang was ranked as the 46th fastest supercomputer in the world supporting Climate, Environment, Electronic and many other science research projects.

Video: Energy Secretary Steven Chu Speaks at Ground Breaking for New LBNL Facility

Rich Brueckner — Fri, 03 Feb 2012 16:06:16 +0000

www.youtube.com/watch?v=kwxxyQauSr0

In this video, Energy Secretary Steven Chu and others speak at the ground-breaking ceremony for the new LBNL Computational Research and Theory (CRT) facility. The CRT will be at the forefront of high-performance supercomputing research and be DOE’s most efficient facility of its kind. Joining Secretary Chu as speakers were Lab Director Paul Alivisatos, UC President Mark Yudof, Office of Science Director Bill Brinkman, and UC Berkeley Chancellor Robert Birgeneau. The festivities were emceed by Associate Lab Director for Computing Sciences, Kathy Yelick.

According to Yelick, the new CRT facility will be one of the most energy efficient supercomputer centers on the planet.

Mexico Powers up “Fire Serpent” Xiuhcoatl Supercomputer

Ralph — Fri, 27 Jan 2012 16:47:42 +0000

Mexico’s Center for Research and Advanced Studies (CINVESTAV) has powered up Xiuhcoatl, a the country’s most powerful supercomputer at 25 Teraflops. Joining a three-way grid knows as Lancad, the system will be used to research such areas as Alzheimer’s, the Earth’s climate, tsunamis, and the formation of stars.

According to Cinvestav chief Rene Asomoza, the Xiuhcoatl (fire serpent in Nahuatl) supercomputer has 3840 cores. As part of the Lancad project, it will generate physical infrastructure that will improve Mexico’s competitive position in the world. Read the Full Story.

Video: Gordon Supercomputer Wows TV Audience

Ralph — Wed, 25 Jan 2012 18:20:02 +0000

In this video from Fox News, researchers describe the power and capabilities of Gordon, the flash-based supercomputer at the San Diego Supercomputer Center.

Video: Sunway Bluelight Super Based on Chinese Processors Starts Operation

Rich Brueckner — Sun, 22 Jan 2012 13:00:36 +0000

Our Video Sunday feature continues with this story announcing that the Chinese Sunway Bluelight supercomputer is now fully operational after a three month trial run in Jinan. Sunway BlueLight is the first supercomputer built entirely of components engineered and built in China.

Bright Cluster Manager Powers Astrophysics at UNIVAP

Rich Brueckner — Wed, 18 Jan 2012 13:00:19 +0000

Today Bright Computing announced that Instituto de Pesquisa & Desenvolvimento — Universidade do Vale do Paraíba (IP&D-UNIVAP) is running Bright Cluster Manager on its new hybrid HPC cluster. Used by UNIVAP’s Astrophysics department to simulate the dynamics of colliding galaxies and cosmological structure formation, the cluster was installed, configured and tested in less than five hours.

We were looking for a cluster management solution that would remove the complexity of managing our CPU/GPU cluster,” said Dr. Irapuan Rodrigues, professor of Physics at UNIVAP. “We are researchers, not sys admins, and prefer to spend our time on science. We chose Bright because it’s efficient, easy to learn and use, and takes little of our time to get the most out of our HPC cluster. Further, we needed a solution that would easily scale as we expand our system. Bright’s unified environment is a big advantage for us.”

Read the Full Story.

Cyprus Unveils Largest Super in Eastern Mediterranean

Ralph — Fri, 13 Jan 2012 16:05:39 +0000

This week the largest supercomputer in the Eastern Mediterranean was unveiled at the Cyprus Institute’s Computation-based Science and Technology Research Centre (CaSToRC). The IBM system is a hybrid CPU/GPU cluster currently holding 1,392 processors in 116 nodes.

The HPC facility “will enable cutting-edge research” the University of Cyprus’ rector Constantinos Christofides said, adding that times of economic crisis was exactly when research and innovation was necessary for growth. The Cy-Tera facility will be serving the research needs of the Cyprus Institute and a host of partners, including the University of Cyprus, Jordan’s Synchrotron-light for Experimental Science and Applications in the Middle East (SESAME), the University of Illinois’ national centre for supercomputing applications, the Julich Supercomputing Centre in Germany, among others.

Read the Full Story.

Video: The Multipurpose KillDevil Cluster

Rich Brueckner — Sun, 01 Jan 2012 17:35:20 +0000

www.youtube.com/watch?v=h5u90aXo7VM

Our Video Sunday feature continues with this SC11 presentation by Dr. Mark Reed from the Scientific Engagement Group for Research Computing at University of North Carolina.

The KillDevil cluster is a Linux-based computing system available to researchers across the campus. With more than 8000 computing cores across 706 servers and a large scratch disk space, it provides an environment that can accommodate many types of computational problems. The blades are interconnected with a high speed Infiniband network, making this especially appropriate for large parallel jobs. Killdevil is a heterogeneous cluster with at least 48 GB of memory per node. In addition, there are nodes with extended memory, extremely large memory, and GPGPU computing (Note: “”KillDevil” is named after the North Carolina coastal town.)

Chinese Sunway Bluelight Super is DEC Alpha, One Decade Removed

Rich Brueckner — Tue, 27 Dec 2011 13:35:26 +0000

Nebojsa Novakovic has posted some interesting details on the Chinese-built processor used in the Sunway Bluelight supercomputer, which ranked number 14 on the recent TOP500 list. The rumors seem to be correct; the CPU is based on the DEC Alpha processor.

SW3 aka SW1600 is a 16-core, 64-bit RISC processor, with each core looking a lot like an improved version of the 21164A EV56 Alpha core, plus vector FP unit extension added to each core. While the initial speed range was 1 to 1.2 GHz in the 65nm process, the standard speed grade is a 1.1 GHz chip with 141 GFLOPs DP FP performance. The speed set for the Bluelight Petaflop machine’s Top 500 run was 975 MHz, though. The quad-channel 128-bit DDR3 on-chip memory controller offers 68 GB/s bandwidth – yes, equivalent to 8 channels of DDR3-1066 server RAM.

Read the Full Story.

Meet Gordon, the 320-terabyte Flash Drive

Rich Brueckner — Tue, 27 Dec 2011 12:45:46 +0000

Tom Pfingsten writes that the $20 Million Gordon supercomputer at SDSC is the biggest flash memory-based system anywhere.

Center spokesman Jan Zverina said Thursday that Gordon is particularly suited for applications such as analyzing traffic patterns and making sense of huge quantities of economic data. “It’s not only fast, but it can really mine through massive, massive amounts of data to find that needle in the haystack —- that little aberration or trend that you want to pick up in reams and reams of data. It’s easily the most powerful supercomputer in Southern California now.”

Read the Full Story.

HokieSpeed Super to become a War Horse for Researchers

Rich Brueckner — Wed, 21 Dec 2011 17:49:00 +0000

Steven Mackay writes that Virginia Tech’s new “HokieSpeed” supercomputer will be a veritable “War Horse” for researchers working on diverse science.

You may remember how Virginia Tech crashed the supercomputing arena in 2003 with System X, a novel Apple server cluster powered by the company’s G5 processors. Ranked at number 96 on the TOP500 and number 11 on the Green500, the new HokieSpeed supercomputer is 22 times faster and yet a quarter of the size of X, with a double-precision peak of 240 teraflops.

HokieSpeed is a versatile heterogeneous supercomputing instrument, where each compute node consists of energy-efficient central-processing units and high-end graphics-processing units,” said Wu Feng, associate professor with the Virginia Tech College of Engineering’s computer science and electrical and computer engineering departments. “This instrument will empower faculty members, students, and staff across disciplines to tackle problems previously viewed as intractable or that required heroic eﬀorts and signiﬁcant domain-speciﬁc expertise to solve.”

Each HokieSpeed node contains two 2.40-gigahertz Intel Xeon E5645 6-core central processing units, and two NVIDIA M2050/C2050 448-core GPUs, which reside on a Supermicro 2026GT0TRF motherboard.

HokieSpeed is now in the final stages of acceptance testing. Read the Full Story.

QLogic Powers World’s Fastest Sandy Bridge Cluster at LLNL

Rich Brueckner — Tue, 13 Dec 2011 14:20:00 +0000

This week QLogic announced that its InfiniBand technologies are powering the world’s fastest cluster powered by Intel’s “Sandy Bridge” Xeon E5 processors. Ranked at #15 on the TOP500, the “Zin” supercomputer at LLNL comprises 46,208 cores in 2916 nodes, producing up to 837 megaflops per watt.

The ever more powerful computing systems Lawrence Livermore requires to fulfill its national security missions must be balanced with increasing energy efficiency,” said Matt Leininger of LLNL’s Advanced Simulation and Computing program. “To meet its scientific computing demands, the Laboratory works with industry leaders to advance HPC.”

I talked to QLogic’s Joe Yaworski about what makes the Zin system so remarkable. He said that the fact that this Top15 cluster achieves nearly a Petaflop with less than 3000 nodes is a testament to the efficiency of the Sandy Bridge processor and QLogic’s TrueScale interconnect.

Yaworski also noted that the Scalable Unit architecture developed by the Tri-Labs enabled the cluster to be deployed very rapidly at the end of October. Using Qlogic’s TrueScale InfiniBand software, LLNL was able to install and configure the cluster in record time and complete the TOP500 benchmark.

Read the Full Story.

Video: Deploying an HPC Cluster: From Breaking Ground to Production Mode of a Hybrid CPU/GPU cluster

Rich Brueckner — Sun, 11 Dec 2011 13:00:24 +0000

www.youtube.com/watch?v=us9TgDACPDM

In this video, Steve Jones of Stanford University presents: Deploying an HPC Cluster: From breaking ground to production mode of a hybrid CPU/GPU cluster. Recorded at the HPC Advisory Council Stanford Workshop on Dec. 6, 2011.

Gordon Flash-Based Super is the Fastest #48 on the Planet

Ralph — Sat, 10 Dec 2011 18:00:17 +0000

Marty Graham from Wired Magazine has posted a profile of SDSC’s Gordon, the world’s first flash-based supercomputer.

Gordon uses 300 terabytes of flash, spanning 1,024 high-performance Intel 710 series drives, and the system includes new software designed to aggregate resources from multiple physical server nodes into “super-nodes,” so users have immediate access to data, rather than waiting for the system to access particular drives. Allan Snavely, the SDSC’s associate director, sees this as the world’s largest thumb drive. Flash memory is stuff used not only in USB thumb drives but cell phones and digital cameras. According to Snavely, Gordon can run massive databases up to 10 times faster than traditional memory, and it now ranks 48th on the official Top500 list of the fastest supercomputer in the world. The project is part of a larger trend in the supercomputer game, where systems are moving away from traditional components, toward new types of hardware that can improve speed, cost, efficiency, and, in the case of the Chinese, independence from the West.

www.youtube.com/watch?v=plFbUIOyIbY

In case you missed it, this video shows SDSC Director Michael Norman and the “IOPSoMeter” at SC11, which clocked the Gordon supercomputer at an unprecedented 35 Million IOPs. Read the Full Story.

RSC Super is Greenest in Russia

Ralph — Fri, 09 Dec 2011 21:04:49 +0000

This week the Russian supercomputer vendor RSC Group announced that their liquid-cooled x86 system installed at South Ural State University is the most energy efficient Russian HPC system according to the Green500 list published in November 2011. The system took the 109th position in the rating. Currently there are only five Russian supercomputers in Green500, their number decreased more than twice compared to the previous edition of the rating.

We are happy that the most powerful supercomputing system built by RSC by this time is a de facto the most energy efficient HPC system in Russia and CIS, as shown in the new edition of Green500 rating. Undoubtedly, our energy efficient RSC Tornado architecture with liquid cooling for widely available standard server boards made a great contribution in this success. Most computing nodes of SKIF Aurora SUSU are based on this architecture”, said Alexey Shmelev, Chief Operating officer at RCS Group.

Read the Full Story.

Video: TACC Expanding to Make Room for Stampede

Rich Brueckner — Thu, 08 Dec 2011 08:37:52 +0000

This week the University of Texas at Austin announced it is funding a $56 million datacenter project to make room for the pending Stampede supercomputer.

In the academic HPC environment, we make a different set of tradeoffs than a business that would require a commercial data center. We’re less concerned about 100 percent reliability and more concerned about the capabilities we can offer when things are running the other 98 percent of the time. We don’t do as much as the commercial data centers in terms of redundancy, but we’re much more aggressive in terms of the amount of power per square foot, for instance.”

The new data center is expected to be complete in July 2012. Read the Full Story.

Video: Penguin Computing Talks APU Supercomputing at SC11

Rich Brueckner — Sun, 27 Nov 2011 13:00:52 +0000

In this video, Penguin Computing CTO Phil Pokorny discusses the company’s latest innovation–an APU-based supercomputer deployed at Sandia National Labs. Recorded at SC11 in Seattle.

Interview: Michael Resch on the Hermit Cray XE6 at HLRS

Rich Brueckner — Tue, 15 Nov 2011 18:00:45 +0000

The new Hermit Cray XE6 system at HLRS in Stuttgart is the third Tier-0 system in the PRACE Research Infrastructure (following JUGENE and CURIE). HLRS has a long history with Cray, so we caught up with Dr. Michael Resch to learn more about the system and its mission.

insideHPC: With it’s vast experience in supercomputing, HLRS undoubtedly had many choices of architectures for this deployment, including building their own. What do you think were the most compelling attributes of the Cray XE6 for this mission?

Michael Resch: HLRS as a ccenter has a focus on engineering and industrial applications. As a consequence sustainable performance was one of the key points in our procurement. Furthermore we wer looking for a partner that could co-develop a next generation system with us and was willing to share knowledge and accept a partnership that goes beyond the usual customer-relationship. Cray was both offering outstanding sustained performance and a special partnership. This is going to be especially important when we will have to discuss GPUs as part of a next generation system.

insideHPC: Can you describe the performance characteristics of the Hermit Cray XE6 supercomputer as compared to its predecessors at HLRS?

Michael Resch: The Cray XE6 already now shows outstanding sustained performance in the Linpack benchmark and outperforms other systems with much higher peak performance. For our applications we expect to see sustained performance to be in the range of 20 – 60 % for our benchmark applications.

insideHPC: I remember my first visit to Stuttgart, attending a Cray User Group meeting back in 1997. What do you think are
the key factors that have driven this long-standing relationship between HLRS and Cray?

Michael Resch: The long standing relationship between Cray and HLRS is based on mutual trust. First and foremost Cray has excellent staff on the ground in Germany and Europe. There are quite some outstanding technical and software experts available that have a clear focus on the customer. Second Cray always understood what the requirements of HLRS were and how they changed over time. Finally Cray accepted the two-vendor policy of HLRS and was not giving up on the account when it was no in the leading position in that two-vendor concept.

insideHPC: How will the European PRACE initiative leverage Hermit as a “Tier-0″ system?

Michael Resch: PRACE users will have full access to the system. They will have to go through all bureaucratic procedures but once these hurdles are taken HLRS and Cray together will do their best to get the most out of the XE6 for every individual European user. Cray ad HLRS have set up a center of competence together which will add several scientific staff members to the crew to get their hands directly on all the codes that target the Petaflop.

insideHPC: The name, “Hermit” would seem to suggest a system that lives in isolation. With it’s mission to to work directly with industrial partners in automotive and aerospace engineering, that is really not the case with this supercomputer, is it?

Michael Resch: Hermit has a number of meanings and I would like to keep it a secret what the name means until we will officially inaugurate the system in spring 2012. Actually the name has a lot to do with Stuttgart and the global change that is currently changing also the global position of the city.

insideHPC: When do you think users will be able to start using Hermit?

Michael Resch: As we speak we start to bring in the first users. We are going through acceptance and this includes early usage. We expect to see general usage to start December 1st.

Cray Lands $188m Blue Waters NCSA Contract, Breaking 10 Petaflops with Opteron-GPU Tag Team

Rich Brueckner — Mon, 14 Nov 2011 09:02:18 +0000

By Timothy Prickett Morgan • Get more from this author

If there was a reason that Cray CEO Peter Ungaro, who formerly ran IBM’s high performance computing biz, was a little extra perky when the company announced its third quarter financials two weeks ago, it was not just because the SC11 supercomputing trade show was coming to Cray’s hometown of Seattle this week. Or that Advanced Micro Devices was once again late with an Opteron launch and had hurt its numbers.

It was because Ungaro knew that the National Science Foundation had ordered the budget for the “Blue Waters” petaflops-scale super to be rejigged and sent out for rebid – and Cray was in the hunt to win what turns out to be a $188m deal.

Bill Kramer, deputy project director of the Blue Waters project at the National Center for Supercomputing Applications at the University of Illinois, tells El Reg that Blue Waters was not a specific system, but rather a complete set of infrastructure, including a data center, plus computation, networking, and storage and, most importantly given the software goals of the NCSA, code that scales to real-world petaflops performance.

“We wanted to change the computational elements, and we got approval after a peer review,” says Kramer.

So out goes IBM’s super-dense Power 775 cluster version of Blue Waters, which Big Blue pulled the plug on back in August because the Power7-based machine was more expensive to manufacture than the company thought when it won the competitive bidding for the project back in 2007.

In comes the largest system that Cray has built in its history – at least until the 10 to 20 petaflops “Titan” ceepie-geepie hybrid that Cray is building for Oak Ridge National Laboratory is installed and if it is fully extended.

Like Titan, the Blue Waters system that Cray is building for NCSA will be a mix of standard XE6 Opteron blade server nodes and XK6 mixed CPU-GPU nodes, all linked together in a single network using the “Gemini” XE interconnect created by Cray. The XE6 blades will be equipped with eight sockets of 16-core “Interlagos” Opteron 6200 processors and the XK6 blades will have four Opteron sockets and one Nvidia GPU coprocessor per blade.

As with Titan, the Blue Waters system that Cray has pitched to NCSA will be based on Nvidia’s next-generation “Kepler” GPUs, which were expected by the end of this year when Nvidia outed its roadmap unexpectedly in September 2010 but which are clearly coming sometime in 2012, not in time for Christmas shopping this year.

The Kepler GPUs will offer a significant performance bump over the current 512-core “Fermi” graphics engines that are used in Nvidia’s GeForce and Quadro video cards and Tesla coprocessors. They are fabbed by Taiwan Semiconductor Manufacturing Corp and use its 28 nanometer processes, which should allow Nvidia to put a lot more cores into a GPU as well as boost the clock speeds a bit. Nvidia hasn’t said much about the Kepler GPU design, but it looks like notebook makers are going to get first stab at them early next year and that performance will be more than double what the Fermi GPUs do today. The top-end Fermi GPUs with all 512 cores running at 1.3GHz can deliver 665 teraflops of double-precision floating point performance.

The specifications of the rebooted Blue Waters system are still a bit in flux, but here’s what NCSA is getting for that $188m. The machine will have at least 235 XE6 cabinets, more than 30 XE6 cabinets, and more than 30 storage and I/O server cabinets. The resulting machine will have more than 49,000 Opteron 6200 processors and more than 380,000 cores, with another 3,000 Nvidia GPU coprocessors packing a hell of a floating point wallop as well. (If the Keplers offer twice the double-precision floating point performance as the Fermis, then those 3,000 GPU coprocessors will account for about 4 petaflops of aggregate oomph.) The plan is to put 4GB of main CPU memory per core into the machine, for a total of more than 1.5PB.

The XE6 and XK6 nodes will all be linked to each other using the XE interconnect through a 3D torus, which will require over 9,000 wires to link it all together. (That’s around 4,500 kilometers of wire if you strung it all out.) The Blue Waters machine is expected to have 11.5 petaflops of aggregate peak number-crunching performance.

Kramer says that the plan is to run the 16-core Opteron 6200s processors with half the cores sleeping a lot of the time and giving the remaining eight cores full access to floating point units that are shared by each integer unit in the Bulldozer modules. By doing this, NCSA will be able to run the cores in a Turbo Core mode that can add anywhere from 600MHz to 1.3GHz of extra clocks to the chip, depending on the Opteron 6200 processor that NCSA chooses for the machine.

“We studied this quite a bit, but for a lot of our applications, it makes sense to run it like an eight-core with dual 128-bit floating point performance,” says Kramer.

Cray is also building a storage subsystem for the Blue Waters machine, which will run the Lustre file system and deliver 25PB of usable disk capacity. It is not clear how that local storage will be linked into the system, that 25PB of capacity will be accessible through more than 1TB/sec of aggregate bandwidth from the cluster. There is an additional 500PB of nearline storage also being added to the machine, which will have 100GB/sec of bandwidth into the cluster. The external network coming into the Blue Waters machine will have 100Gb/sec of bandwidth at first, and will eventually scale up to 300Gb/sec.

The Blue Waters machine will run the Cray Linux Environment, the company’s tweak on SUSE Linux 11 that also has an Ethernet network compatibility mode that allows applications compiled for Linux machines clustered using Ethernet to run unmodified and in emulation mode on top of the XE interconnect.

The Blue Waters machine will be delivered by Cray in phases over the next six to nine months, with the initial delivery early next year and early program deployment by the middle of 2012. The plan is to have the full machine operational by the end of 2012, which is more or less the same timing that NCSA expected with the IBM Power7 Blue Waters behemoth. ®

This article originally appeared in The Register. It appears here in its entirety as part of a cross-publishing agreement.

Video: A Closer Look at China’s Homegrown Sunway BlueLight Super

Rich Brueckner — Sun, 06 Nov 2011 13:00:06 +0000

www.youtube.com/watch?v=MaECfRDsTOQ

In this video, we get a closer look at the components of the Sunway BlueLight supercomputer in China. The recently announced Petaflop machine is based on a Chinese 16-core processor, rumored to be a derivative of the DEC Alpha 21164. Sunway BlueLight is expected to come in somewhere around number 20 in the pending TOP500 list, which is expected to be release on Nov. 15, 2011.