insideHPC

Podcast: Radio Free HPC Interviews Jack Dongarra on the New TOP500

Rich Brueckner — Mon, 17 Jun 2013 18:21:59 +0000

In this podcast, the Radio Free HPC team discusses the June 2013 TOP500 list with Jack Dongarra from the University of Tennessee. How did the Chinese manage to trump the world once again with the #1 supercomputer on the planet? Listen in to find out.

Download the MP3 * Subscribe on iTunes * RSS Feed

The post Podcast: Radio Free HPC Interviews Jack Dongarra on the New TOP500 appeared first on insideHPC.

Excitement at ISC13: Getting to the MilkyWay 2

johnH — Mon, 17 Jun 2013 08:40:51 +0000

In this special guest feature, Intel’s John Hengeveld describes the effort that went into building the fastest supercomputer on Earth.

Milky Way 2 is amazing.. and whats more amazing is… you can build a similar system for your needs.

So by now you have all heard about the MilkyWay 2 system in China that has surprised the world and achieved the #1 spot on the TOP500 list. The Intel team has been working on many fronts in the past year. The upcoming Intel Xeon Processors E5-2600 V2, a major expansion to the Intel Xeon Phi products family, expanding our industry leading software development toolkits. We’re in like 98% of the new systems on the TOP500 list.. so seriously.. we need a rest.

All of these things are coming together at ISC13. The outgrowth of much of this work is shows up in the MilkyWay 2 system, but what’s more important is that this technology is very soon available from a broad range of suppliers to be turned on a wide range of industrial and scientific technical computing applications.

The Processor and Coprocessor components Intel proposed and shipped in production for MilkyWay2 are being announced and demonstrated by Intel for the first time at ISC today. The first demonstration of the as yet unlaunched Intel Xeon Processor E5-2600 V2 product family will be shown in the Intel Booth in a live 52 node cluster showing high fidelity visualizations of an Audi RS5 vehicle design.

Later this afternoon, Raj Hazra will announce the general availability of 5 new Intel Xeon Phi Coprocessor products including the 3100 family that is featured in the MilkyWay 2 system. Please check out Raj’s keynote at 6pm Leipzig time on insidehpc.com for the full story live when it happens.

I want to tell a short story about the MilkyWay 2 system that you probably haven’t heard, and show why being at Intel is the coolest thing ever.

About a year ago, the NUDT folks, led by Professor Liao, had a really good idea of how he wanted to build the worlds biggest super computer, using the next step in their proprietary fabric and trying to use intel’s latest processors to achieve its objectives in power efficiency and performance.

They gave Intel and others programmability, node power and node performance goals with very tight constraints. Then they dropped the heavy challenge… “and it all has to work and be #1 in the world by June 2013”. Intel proposed a solution that not only met all their time and performance requirements but also their programmability requirements.

What they wanted they later described as a “Neo-Heterogeneous Architecture”. This type of system has two tiers of hardware heterogeneity, but driven off a consistent programming model and parallelism abstraction. This allows very much faster development of applications that scale to a very high level.

What the use of Intel Xeon Phi Coprocessors offered is hardware with the performance and energy efficiency required, but removed the need for an alternate programming model for the second tier thus enabling their neo-heterogeneous architectural vision.

The plan to build out the system as audacious as well. The customer developed the system based on Intel Xeon Processor E5-2600 based products and built test cluster that would itself be on the top500 list. After debugging that system and the code that ran on it, NUDT planned to get the next generation product (the future Intel Xeon Processor E5-2600 V2 family) and basically drop it in. This enabled them to go from first delivery of a blade to a completed system in about 4 weeks. The Linpack run came about a week later. By the time Jack Dongarra and others saw it at the end of May, it was running real applications.

For Intel’s part, what Intel’s factory and engineering teams did was validate two new products and put them into high volume production on schedule to within a week of our schedule predicted a year before.

I have been getting asked why Intel got chosen for this amazing system. The short answer is.. because we deliver. We delivered.

The post Excitement at ISC13: Getting to the MilkyWay 2 appeared first on insideHPC.

It’s Official: China Once Again Leads the TOP500

Rich Brueckner — Mon, 17 Jun 2013 08:39:18 +0000

The June 2013 TOP500 was released today at ISC’13 with an all-new Chinese supercomputer leading the list with 33.86 Petaflops on the Linpack benchmark.

Tianhe-2, or Milky Way-2, will be deployed at the National Supercomputer Center in Guangzho, China, by the end of the year. The surprise appearance of Tianhe-2, two years ahead of the expected deployment, marks China’s first return to the No. 1 position since November 2010, when Tianhe-1A was the top system. Tianhe-2 has 16,000 nodes, each with two Intel Xeon IvyBridge processors and three Xeon Phi processors for a combined total of 3,120,000 computing cores.

Formerly number one on the list, the Titan Cray XK7 at Oak Ridge is now ranked No. 2 with 17.59 Petaflops and Sequoia, an IBM BlueGene/Q system installed at LLNL falls to number 3 on the list with 17.17 petaflops.

Other highlights from the June 2013 TOP500 list include:

There are 26 systems with performance greater than a Petaflop, up from 23 six months ago.
The new No. 1 system, Tianhe-2, and the No. 6 system, Stampede, are using Intel Xeon Phi processors to speed up their computational rate. The No. 2 system, Titan, and the No. 10 system, Tianhe-1A, are using NVIDIA GPUs to accelerate computation.
A total of 54 systems on the list are using accelerator/co-processor technology, down from 62 in November 2012. Thirty-nine of these use NVIDIA chips, three use ATI Radeon, and eleven systems use Intel MIC technology (Xeon Phi).
The number of systems installed in China has now stabilized at 66, with 72 and 68 on the last two lists. As a nation, China now holds the No. 2 position as a user of HPC, ahead of Japan, UK, France, and Germany. Due to Tianhe-2, China has also taken the No. 2 position in the performance share, ahead of Japan.
Intel continues to provide the processors for the largest share (80.4 percent) of TOP500 systems.

Read the Full Story.

The post It’s Official: China Once Again Leads the TOP500 appeared first on insideHPC.

nCore HPC Rolls Out BrownDwarf ARM DSP Supercomputer

Rich Brueckner — Mon, 17 Jun 2013 08:07:28 +0000

We don’t get the chance to announce new players in the HPC space very often, especially from our hometown of Portland. Today newcomers nCore HPC announced the BrownDwarf Y-class supercomputer, a heterogeneous ARM- and DSP-based system designed for green high performance computing.

With its unique parallel computing architecture and a high performance, low latency interconnect, the BrownDwarf is “Y-Class supercomputer” with extremely low power consumption. In a 144 node configuration, BrownDwarf delivers 70 Teraflops of performance at 10kw inside a 42U high rack.

The BrownDwarf Y-Class system is an incredibly important milestone in HPC system development,” said Ian Lintault, managing director of nCore HPC. “Working in close collaboration with TI, IDT and our hardware partner Prodrive, we have successfully established a new class of energy efficient supercomputers designed to fulfill the demands of a wide range of scientific, technical and commercial applications.”

The BrownDwarf Y-Class node leverages multiple Keystone-II and Keystone-I SoC components from Texas Instruments Incorporated (TI). Each node integrates four ARM Cortex-A15 MPCore processors, 24 TMS320C66x digital signal processor (DSP) cores and 26GB of ECC memory.

Read the Full Story.

The post nCore HPC Rolls Out BrownDwarf ARM DSP Supercomputer appeared first on insideHPC.

C-DAC Initiatives in HPC for Scientific & Engineering Research

Rich Brueckner — Sun, 16 Jun 2013 15:11:13 +0000

In this video from the HPC Advisory Council European Conference 2013, Abishek Das from C-DAC presents: C-DAC Initiatives in HPC for Scientific & Engineering Research.

Exploitation of parallel processing technologies for Scientific & Engineering Research will accelerate research of applications in various engineering disciplines that employ HPC techniques and facilitates research about HPC technologies contributing to capacity-building of the country.”

Download the slides (PDF).

The post C-DAC Initiatives in HPC for Scientific & Engineering Research appeared first on insideHPC.

Gordon Supercomputer Fosters New Fields of Research for HPC

Ralph — Sun, 16 Jun 2013 15:04:20 +0000

With over 300 TB of high performance Intel flash memory SSD, the Gordon supercomputer at SDSC is opening the door to new areas of research for HPC such as political science, mathematical anthropology, finance, and even the cinematic arts.

Gordon’s extraordinary speed makes it possible for researchers to tackle questions they couldn’t address before, simply because they didn’t have a system that was uniquely tailored to handle the challenges of data intensive computing,” said SDSC Michael Norman just prior to Gordon’s launch. “I view Gordon as a new kind of vessel, a ship that will take us on new voyages to makes new discoveries in new areas of science.”

Read the Full Story.

The post Gordon Supercomputer Fosters New Fields of Research for HPC appeared first on insideHPC.

Video: Xyratex Update – ClusterStor

Rich Brueckner — Sun, 16 Jun 2013 14:11:32 +0000

In this video from the HPC Advisory Council European Conference 2013, Torben Kling-Petersen from Xyrtex presents: Xyratex Update – ClusterStor.

Download the slides (PDF).

The post Video: Xyratex Update – ClusterStor appeared first on insideHPC.

Hungarian Datacenter Inaugurated

Staff — Sun, 16 Jun 2013 13:47:42 +0000

CERN and the Wigner Research Centre for Physics have inaugurated the Hungarian data centre in Budapest, marking the completion of the facility hosting the extension for CERN computing resources.

About 500 servers, 20,000 computing cores, and 5.5 Petabytes of storage are already operational at the site. The dedicated and redundant 100 Gbit/s circuits connecting the two sites are functional since February 2013 and are among the first transnational links at this distance. The capacity at Wigner will be remotely managed from CERN, substantially extending the capabilities of the Worldwide LHC Computing Grid (WLCG) Tier-0 activities and bolstering CERN’s infrastructure business continuity.

WLCG’s mission is to provide global computing resources to store, distribute and analyse more than 25 Petabytes of data annually generated by the Large Hadron Collider (LHC). It is a global system organised in tiers, with the central hub being the Tier-0 at CERN.

The experiments” computing resources needs will increase significantly when the LHC restarts in 2015. Hosting computing equipment at the Wigner Centre to extend CERN’s data centre Tier-0 capabilities is essential for dealing with this expected increase, and to the success of our physics programme. The remote capacity will also contribute to business continuity for the critical systems in case of a major issue on CERN’s site,’ said CERN director-general Rolf Heuer. “A number of sciences currently face exponential data growth. This innovative approach with Wigner could point the way for research centres to run their services in the future.”

This story appears here as part of a cross-publishing agreement with Scientific Computing World.

The post Hungarian Datacenter Inaugurated appeared first on insideHPC.

Video: High Performance Computing Trends for 2013

Rich Brueckner — Sun, 16 Jun 2013 13:36:56 +0000

In this video from the HPC Advisory Council European Conference 2013, Addison Snell from Intersect360 Research presents: High Performance Computing Trends for 2013.

This presentation is an overview of the current important trends in HPC, based on the latest end-user research studies and market forecasts. Topics include accelerator adoption, the role of HPC in Big Data, and the ratio of spending between hardware, software, staffing, and facilities.”

Download the slides (PDF).

The post Video: High Performance Computing Trends for 2013 appeared first on insideHPC.

Livermore to Share 5 Petaflops with Industry

Ralph — Sun, 16 Jun 2013 12:16:13 +0000

Over at FCW Magazine, Mark Rockwell writes that Lawrence Livermore National Labs has five petaflops of computing power it wants to share. To help fuel technological innovation to spur U.S. economic competitiveness, the research facility has officially opened up access to its massive Vulcan supercomputer to industry and academia for collaborative developmental projects.

High performance computing is a key to accelerating the technological innovation that underpins U.S. economic vitality and global competitiveness,” said Fred Streitz, HPC Innovation Center director, in a June 11 statement. “Vulcan offers a level of computing that is transformational, enabling the design and execution of studies that were previously impossible, opening opportunities for new scientific discoveries and breakthrough results for American industries.”

During its initial shakeout period, LLNL said Vulcan was combined with the larger Sequoia system to produce set a world speed record of 504 billion events per second for a discrete event simulation in collaboration with the Rensselaer Polytechnic Institute. That achievement, said the lab, opens the way for the scientific exploration of complex, planetary-sized systems.

Read the Full Story.

The post Livermore to Share 5 Petaflops with Industry appeared first on insideHPC.

Video: Efficient and Cost-Effective HPC Interconnection Networks

Rich Brueckner — Sun, 16 Jun 2013 10:48:58 +0000

In this video from the HPC Advisory Council European Conference 2013, Jesús Escudero, from the Universidad de Castilla-La Mancha presents: Efficient and Cost-Effective HPC Interconnection Networks.

This keynote address analyzes the most prominent challenges for designing and cost-effective interconnection networks for Exascale systems, such as topology scalability, power consumption, fault tolerance, and/or congestion control. Besides, some solutions are proposed and their implementation complexity in commercial products are estimated.”

The post Video: Efficient and Cost-Effective HPC Interconnection Networks appeared first on insideHPC.

TACC to Deploy 20 Petabyte Global File System from DDN

Ralph — Sun, 16 Jun 2013 10:47:22 +0000

This week TACC announced it will deploy 20 Petabyte, scalable global file system (GFS) from DDN that will be accessible to all of TACC’s computing and visualization systems.

The rate of data growth during the last decade has been exponential,” said Tommy Minyard, TACC’s director of Advanced Computing Systems. “This new global file system will offer a very large storage pool for thousands of researchers to store persistent data and make it easier for them to perform their scientific research. They will be able to access data from any production system at TACC and use it for later analysis, further processing, or to continue to run their applications.”

The new GFS will provide more than 20 petabytes of storage capacity for scientific data, and will allow scientists to access their data rapidly through its aggregate bandwidth of greater than 100 gigabytes per second. By providing a massive storage capability with high-performance access from all of TACC’s production systems, the system will further diversify the resources available to researchers and eliminate the need for researchers to manage data across multiple systems. This will balance the storage available to all TACC systems and make migration between systems and upgrades to new ones more transparent.

Read the Full Story.

The post TACC to Deploy 20 Petabyte Global File System from DDN appeared first on insideHPC.

Video: Scalable MPI Futures

Rich Brueckner — Sun, 16 Jun 2013 10:41:19 +0000

In this video from the HPC Advisory Council European Conference 2013, Richard Graham from the MPI Forum presents: Scalable MPI Futures.

Download the slides (PDF).

The post Video: Scalable MPI Futures appeared first on insideHPC.

MEGWARE Sponsors Local Team at ISC’13 Student Cluster Challenge

Rich Brueckner — Sat, 15 Jun 2013 11:30:20 +0000

When it comes to Student Cluster Competitions, is there such a thing as a home team advantage? A team of six Chemnitz students will battle it out with eight other teams in the ISC’13 Student Cluster Challenge. The team calls themselves “TurboTUC” and their roster includes some students with pure science backgrounds in areas such as chemistry and theoretical physics to go along with their computer-related smarts.

The challenge for us is firstly to create high-performance hardware and then use the right components,” says team member Sebastian Siegert. “In addition, we will only find out about part of the applications at the fair itself – then the task will be to discover whether – and how well – we can adapt these with our supercomputer in order to solve the problem as quickly and with the greatest energy efficiency as possible,” adds Henrik Kretzschmar.

Read the Full Story or check out more Team profiles at the Student Cluster Competition Blog.

The post MEGWARE Sponsors Local Team at ISC’13 Student Cluster Challenge appeared first on insideHPC.

New APAX HDF Library Accelerates HDF-Enabled Applications 3-8X

Rich Brueckner — Sat, 15 Jun 2013 11:00:06 +0000

This week Samplify released its new APAX HDF (Hierarchical Data Format ) Storage Library for HPC, Big Data, and cloud computing applications.

HDF is an open source library which is the standard file format for the interchange of climate data. In its current release HDF5 version 1.8.11, the HDF Group adds support for third party plug-ins into the storage pipeline. Samplify’s APAX plug-in for HDF enables transparent access to APAX compression technology without requiring modification of the application software. When used in conjunction with Samplify’s APAX Profiler, the encoding rate can be optimized for each dataset being stored in the file.

Our engagements with government labs, academic institutions, and private data centers reveal a continuous struggle to manage an ever increasing amount of data,” says Al Wegener, Founder and CTO of Samplify. “We have been asked for a simpler way to integrate our APAX encoding technology in Big Data and cloud applications. By using plug-in technology for HDF, we enable any application that currently uses HDF as its storage format to get the benefits of improved disk throughput and reduced storage requirements afforded by APAX.”

At ISC’13 next week, Prof. Dr. Thomas Ludwig will describe experiences with the new Samplify technology in a session entitled: Combining HPC & Big Data in the Climate Modeling Workflow.

Read the Full Story.

The post New APAX HDF Library Accelerates HDF-Enabled Applications 3-8X appeared first on insideHPC.

DDN Hosts User Group Meeting at ISC’13 June 17

Rich Brueckner — Fri, 14 Jun 2013 15:31:29 +0000

The annual DDN User Group meeting will be held on Monday, June 17 in conjunction with ISC’13.

DDN European User Group

Date: June 17, 2013
Time: 8:30am – 11:00am
Place: Congress Center Leipzig, Room M05

Speakers:

University College of London: Ms. Clare Gryce (Head of Research Computing) and Dr. Daniel Hanlon (Research Data Storage Architect)
CSC Finland: Pietari Hyvarinen (System Specialist)
Karlsruhe Institute of Technology – KIT: Jos van Wezel (Storage Group Leader)
DataDirect Networks: Mr. Dave Fellinger (Chief Scientist), Dr. James Coomer (Sr. Technical Advisor), and Mr. Jeff Denworth (VP of Marketing)

In related news, be sure to check out our feature story on the full spectrum of ISC’13 Ancillary Events.

The post DDN Hosts User Group Meeting at ISC’13 June 17 appeared first on insideHPC.

Titan Completes Acceptance Testing

Rich Brueckner — Fri, 14 Jun 2013 11:00:47 +0000

This week the Titan supercomputer at ORNL completed rigorous acceptance testing to ensure the functionality, performance and stability of the machine, one of the world’s most powerful supercomputing systems for open science. As the fastest supercomputer in the world in the November 2012 TOP500, Titan is a Cray XK7 supercomputer comprises 18,688 NVIDIA Tesla GPUs, 299,008 AMD Opteron CPU cores, and 710 terabytes of memory.

The real measure of a system like Titan is how it handles working scientific applications and critical scientific problems,” said Buddy Bland, project director at the Oak Ridge Leadership Computing Facility. “The purpose of Titan’s incredible power is to advance science, and the system has already shown its abilities on a range of important applications and has validated ORNL’s decision to rely on GPU accelerators.”

According to reports in KnoxNews, the Titan system acceptance was delayed due to a problem with faulty solder connections at the pc board level, which required extensive rework by Cray. Read the Full Story.

The post Titan Completes Acceptance Testing appeared first on insideHPC.

STFC to Develop Hybrid Supercomputer Applications

Ralph — Fri, 14 Jun 2013 10:30:04 +0000

The Science and Technology Facilities Council (STFC) at Daresbury in the U.K. has entered an agreement with Nvidia to help develop software for next-generation hybrid supercomputers.

This agreement combines NVIDIA’s leading-edge GPU accelerator technologies and HPC expertise with STFC’s software development expertise,” said David Corney, acting director of STFC’s Department of Scientific Computing. “This unique combination will enable the development of next-generation massively parallel applications, which will be used for exascale performance levels, or a thousand times more powerful than Blue Joule at STFC, the most powerful computer in the UK today.”

The collaboration will offer scientists access to one of the largest software development laboratories in the world, STFC’s Hartree Centre at Daresbury Laboratory, which is dedicated to modelling and simulation software, as well as to Nvidia’s expertise. Read the Full Story.

The post STFC to Develop Hybrid Supercomputer Applications appeared first on insideHPC.

E4 to Showcase Kayla GPU-ARM Development Platform at ISC’13

Ralph — Fri, 14 Jun 2013 10:00:41 +0000

Nvidia made waves in March with the announcement of the Kayla development platform for CUDA on ARM processors. Next week at ISC’13 in Germany, E4 Computer Engineering will demonstrate its new, low-power E4 ARKA SERIES, a Kayla device equipped with an Nvidia Tegra Quad-Core ARM A9 CPU as well as an Nvidia GPU.

For the datacenter, E4 will also demonstrate the ARKA EK002, a 3U dual-node server featuring Nvidia Tegra3 CPU QUAD-Core ARM and a Kepler K20 GPUs configured with a Mellanox ConnectX-3 FDR 56Gb/s InfiniBand Adapter. The device is designed to be a development platform for HPC, Finance, and Energy applications.

To learn more, check out the E4 booth #267 at ISC’13.

The post E4 to Showcase Kayla GPU-ARM Development Platform at ISC’13 appeared first on insideHPC.

ClusterVision Pumps Up Supercomputer Muscles in Brussels

Staff — Thu, 13 Jun 2013 19:00:41 +0000

ClusterVision has announced the successful completion of a new cluster installation project at the Université Libre de Bruxelles (ULB). With cluster design, build and service projects already completed at the Université Catholique de Louvain, and the Université de Liège, the new installation at the Université Libre de Bruxelles continues ClusterVision’s recent history of engagement with universities in the Wallonia region, and also its successful partnership with Dell and others.

The new CÉCI cluster to be installed at the ULB had to feature a very high number of cores interconnected with an InfiniBand network and with a well-balanced storage system to allow fast IO response to the CPU load while enabling high capacity expansions,” said Raphael Leplae ULB, Computing Centre, operations/HPC manager. “ClusterVision proposed a unique combination of hardware and software components matching exactly our needs. This new cluster installation will enable researchers requiring HPC solutions, such as earth-level simulations, bioinformatics and other big data-related analysis, to move towards more ambitious projects and/or be more competitive at the international level.”

ClusterVision worked with system engineers at ULB to design and install a combination of HPC technologies from partners including Dell, Supermicro, AMD, Nvidia, Qlogic, GPFS, and Bright Computing. The compute density and performance characteristics are achieved through the addition of 42 Dell PowerEdge M915 Blade Servers, each with four AMD Opteron 6272 CPUs and 256GB of RAM. Selecting the AMD Opteron 6200 series ‘Interlagos’ 16-core processors allowed engineers at ClusterVision to achieve the scaled performance but keeping within the requirements for density and energy efficiency. Additional compute power is provided by two Nvidia Tesla M2090 GPUs, housed in a Supermicro chassis, also with AMD Opteron 6200 processors.

The system incorporates both high- and mid-range storage expansion based on Dell PowerVault. High-range storage is expanded by Dell PowerVault MD3220 with 24x 900GB SAS 10K RPM. The mid-range storage expansion of 40TB is achieved via one Dell PowerVault MD3200 and two Dell PowerVault MD1200 units. Network connectivity is primarily provided by Qlogic Infinipath QLE7340 and Qlogic QLA12200 switches.

To complete the system, ClusterVision included a full software environment, including IBM’s high performance shared file system, GPFS (General Parallel File System) – an innovative solution for non-IBM storage, and Advanced Version licences of Bright Cluster Manager from Bright Computing. Bright Cluster Manager was used for the installed Linux environment, initial provisioning and configuration of the cluster components and is also the base platform for the day-to-day operational management of the system.

This story appears here as part of a cross-publishing agreement with Scientific Computing World.

The post ClusterVision Pumps Up Supercomputer Muscles in Brussels appeared first on insideHPC.