Entries filed under “ISC12”

Video: Mike Stolz on the New ClusterStor 6000 for HPC

In this video, Mike Stolz from Xyratex describes the company’s new ClusterStor 6000 product and why Xyratex is leading the way in HPC storage performance without sacrificing reliability.

The ClusterStor product family of scale-out HPC data storage solutions features a tightly integrated distributed computing architecture. Xyratex has integrated Lustre into storage appliances designed to operate as turnkey solutions, contrasted with the traditional piecemeal, less-productive HPC storage solutions. The integration of heterogeneous traditional nodes into high performance, dense platforms results in a reduced footprint and the foundation for lower cooling and infrastructure costs,” said Steve Conway, Research Vice President, High Performance Computing, IDC. “Xyratex is involved in some of the largest and most challenging storage deployments in the world, at least one of which aims to become the first high-density platform to scale to 1 TB/second sustained file system performance.”

Recorded at ISC’12 in Hamburg.


Also posted in Events, HPC, HPC Hardware, Storage, Video | Leave a comment

Video: For HPC Energy Efficiency, Samsung 20-nm Memory Fab Makes a Difference

In this video, Peyman Blumstengel from Samsung describes the reasons why the fabrication process used by your system memory vendor really matters for HPC clusters in terms of energy efficiency, performance, and sustainability. Recorded at ISC’12 in Hamburg.

Samsung’s latest 20-nm class DDR3 memory solutions enable power savings of 67% for the memory, as compared to the existing 50-nm class DDR3 solutions. When deployed in a standard server with 96GB of memory (commonly referred to as a 96GB server), this translates into huge savings.

Read the Full Story.

Also posted in Events, Green HPC, HPC, HPC Hardware, Video | Leave a comment

NetApp Tackles Huge IO Storage Requirements for #1 Sequoia Supercomputer

In this video, Bryan Berezdivin from NetApp describes the company’s recent storage deployment for the Sequoia supercomputer at LLNL. Currently ranked at #1 on the TOP500, the 16.3 Petaflop Sequoia system had massive IO requirements of over 1 Terabyte per second served up by the NetApp HPC Solution for Lustre. The total NetApp system storage pool has over 55 Petabytes capacity.

Recorded at ISC’12 in Hamburg. Read the Full Story.

Also posted in Events, HPC, HPC Hardware, New Installations, Storage, Video | Leave a comment

Video: New Accelerations for Parallel Programming

In this video, Todd Wilde from Mellanox presents: New Accelerations for Parallel Programming. Recorded at the HPC Advisory Council Workshop at ISC’12 in Hamburg. Download the slides (PDF).

Also posted in Events, HPC, HPC Hardware, HPC Software, Network, Video | Leave a comment

Video: Spectra Logic Brings High Performance to Big Data Tape at ISC’12

In this video. Molly Rector from Spectra Logic describes the company’s high-performance, Big Data tape storage solutions for the HPC market. Recorded at ISC’12 in Hamburg.

In related news, Spectra recently announced that it will be the sole provider of tape storage for the 10 Petaflop Blue Waters supercomputer project at NCSA.

We are pleased to partner with NCSA and support one of the most powerful and cutting-edge supercomputers in the world,” said Nathan Thompson, founder and CEO, Spectra Logic. “Spectra Logic has been a constant innovator of tape technologies over the past thirty years. It is gratifying to see tape-based storage play a major role in one of the largest, best practice HPC deployments to date and to help support the important scientific breakthroughs and advancements the Blue Waters project will enable.”

Read the Full Story.


Also posted in Events, HPC, HPC Hardware, Storage, Video | Leave a comment

Video: Gnodal Steps Up with 40 GigE Solutions for Extreme Scalability

In this video, Gnodal CTO Fred Homewood describes the company’s high bandwidth, low latency 40 GigE Ethernet switch solutions for HPC and the Enterprise.

The GS0072 is an order of magnitude faster than most competitive switches with congestion management and dynamic load-balancing, allowing the switches to operate at near 100% bandwidth utilization (throughput). The GS0072 is targeted at Cloud and Big Data environments where 10GbE edge connectivity is required to accommodate very large numbers of virtual clients as well as High Performance Computing applications that may require native 40 Gbps capability to achieve extreme scalability and throughput.”

Recorded at ISC’12 in Hamburg. Read the Full Story.

Also posted in Events, HPC, HPC Hardware, Network, Video | Leave a comment

Video: ScaleMP to Support Intel Xeon Phi Coprocessors

In this video, ScaleMP CEO Shai Fultheim describes how vSMP Foundation supports the new Intel Xeon Phi coprocessor. Recorded at ISC’12 in Hamburg.

vSMP Foundation for the Intel Xeon Phi coprocessor based platform will virtualize both the Intel Xeon processor and the Intel Xeon Phi coprocessor based platform cores as well as host and Intel Xeon Phi coprocessor memory to act like a single Symmetric Multiprocessing (SMP) system. Users will have easy access to all of the computing and memory resources, greatly simplifying their development and production environments,” stated Shai Fultheim, founder and CEO of ScaleMP.

Read the Full Story.

Also posted in Accelerators, Events, HPC, HPC Hardware, HPC Software, Video | Leave a comment

From Garage-Shop Roots, MEGWARE Leads German TOP500 Vendors

We hear a lot about the TOP500 this time of year out of ISC Hamburg, but you may be wondering how the German supercomputing fared this time. It looks like MEGWARE out of Chemnitz is the leader with three projects in the TOP500: ABEL, MOGON and VSC-2.

The HPC-Cluster MOGON, which was installed at the University of Mainz in the past weeks, is in place #81 with 33,792 processor cores and 204.99 Tflop/s. It takes 8th place among the German HPC systems. The ABEL-Cluster at the University of Oslo, which MEGWARE also installed in the past weeks, is ranked #96 with its 10,080 cores and 178.6 Tflop/s. The third MEGWARE system is the VSC-2. Austriaís most powerful supercomputer has been in operation at the Technical University of Vienna since May 2011. Last year it got started in 56th place. Currently, it is in 117th place.

Read the Full Story.

Also posted in Business of HPC, Events, HPC | Leave a comment

Mark Seager Presents a Vision for Exascale

In this video, Mark Seager, CTO of the HPC Ecosystem at Intel presents: A Vision for Exascale. Recorded at ISC’12 in Hamburg.


Also posted in Events, Exascale, HPC, Video | Leave a comment

Video: HPC Analyst Crossfire from ISC’12

In this video from the closing session of ISC’12, Addison Snell of Intersect360 Research moderates Analyst Crossfire. Customer panelists were Dona Crawford from LLNL and Happy Sithole from the Center for High Performance Computing in South Africa. On the vendor side was Elena Churakova from T-Platforms and Dave Turek from IBM.

This was the second year that ISC wrapped with a Crossfire session, and I think it just continues to get better. I came up with the idea at ISC’10 and it was great to sit back and watch a professional video crew record the thing in front of a live audience.

Also posted in Events, HPC, Video | Leave a comment

ISC’12 Wraps Up with Record Attendance and a Little Rock and Roll

ISC’12 wrapped up today with news that the conference had record turnout with 2403 attendees.

Even though we keep setting attendance records each year, it’s exciting when we reach or projected goal and having 2,403 attendees this year was very gratifying,” said ISC Executive Director Martin Meuer. “It’s also exciting to see so many members of the HPC community in Asia making the trip to Germany to join us.”

And while we all know that supercomputing is serious business, at ISC, they surely know how to show attendees a good time as you can see in this video. Next year the show moves to Leipzig. Now, don’t youto want go?

Read the Full Story.

Also posted in Events, HPC | Leave a comment

Meet the Winners of the ISC’12 Student Cluster Challenge

To wrap up ISC’12, China won the day with at the Student Cluster Competition. Team Tsinghua did a terrific job, deriving an unprecedented 2.7 Teraflops from their Inspur machine on a meager diet of only 3000 Watts!

In this video, Dan Olds from Gabriel Consulting profiles the winners of the ISC’12 Student Cluster Competition.

Chinese universities have gone student cluster crazy over the past year. When China was allotted two team slots at the ISC 2012 Student Cluster Competition, more than 300 schools expressed interest in participating. Thirty submitted proposals, forcing the country to conduct a play-in round to select two teams to make the Hamburg trip. The overall winner was Beijing’s Tsinghua University (profiled here). They’re a bit of an enigma, since they’re a first-time participant on the world Student Cluster Competition stage. But they beat out NUDT in the intra-China competition, and that means that they know their way around a cluster.

Be sure to check out Dan’s full coverage of the competition including profiles of all the young teams. These kids are all winners in my book, and the HPC skills they’re building will pave the way for great careers in our community when they graduate.

Also posted in Events, HPC, Video | Leave a comment

Interview: Supermicro Shines at ISC’12

Supermicro always impresses me at HPC shows with a wide array of hardware on display. They must be targeting us folks in the geek crowd, because no other vendor has more components, motherboards, and form factors in their exhibit. To learn more, I caught up with David Okada, the company’s Senior Marketing Manager.

insideHPC: Supermicro always seems to always be among the first-to-market with shipping products after new processor and GPU announcements. Does this kind of agility give you a competitive advantage?

David Okada: Supermicro works very closely with our technology partners through development stages to product launch. Even as we extend and evolve our own product lines, we are constantly qualifying future technologies and preparing our platforms for next generation products. This certainly gives us a competitive advantage as we offer the widest range of Server Building Block Solutions ready for the market at any given launch and Supermicro’s customers benefit with fastest access to leading edge technology with not just one or two choices, but a wide range of options so they can create solutions to maximize their margins.

insideHPC: How important is energy efficiency to Supermicro and its customers?

David Okada: Energy-efficiency is a major driving factor for Supermicro’s product development. As we provide the computing infrastructure for large scale data centers, it is our corporate responsibility to ensure we do our part to reduce the power consumption and improve the green computing profile of our solutions. We engineer our own advanced high efficiency (95%+) digital switching, Platinum Level power supplies and Battery Backup Power (BBP™) modules to best manage the power delivery to our solutions. We also specialize in board and chassis level design to optimize component layout and cabling for the most efficient airflow which reduces fan speed and resulting power consumption. We also innovate with high density architectures such as our Twin, SuperBlade® and Double-Sided Storage, MicroCloud™ and most recently launched FatTwin™ platforms to share power and cooling resources further reducing material waste and maximizing power utilization. Our latest push is to combine this power and thermal management expertise to offer free-air cooling solutions that operate in high ambient temperature environments up to 47°C effectively eliminating air conditioning costs and power consumption. We feel confident that customers utilizing complete PUE (power usage effectiveness) optimized solutions from Supermicro will not only lower their TCO, but will also help the planet and our environment in the process.

insideHPC: Does Supermicro have much play in the TOP500, or is that not your sweet spot?

David Okada: Yes, Supermicro has the widest range of GPU server systems on the market and is very experienced with applying GPU technology in supercomputer clusters. We are featured on the TOP500 with a variety of deployments and will be growing our presence with new scalable architectures such as our recently released FatTwin™ platform.

insideHPC: ISC will feature 170 exhibits. What makes Supermicro stand out in the HPC marketplace?

David Okada: Supermicro has launched a new high-density compute/high-capacity storage platform, FatTwin™ (www.supermicro.com/FatTwin), that supports dual 135W Intel® Xeon® E5-2600 processors and up to 8 hot-swappable 3.5″ HDDs per U in 8/4/2 node configurations. These systems are also PUE optimized to operate in high ambient temperature environments up to 47°C. With Supermicro’s FatTwin, Data Centers can implement high-performance, highly-efficient, free-air cooled solutions in industry standard 19″ rack configurations that can be flexibly configured in multiple ways to solve a wide variety of customer needs..

insideHPC: What will Supermicro be showcasing at ISC’12?

We are excited about all of our new products including the Supermicro FatTwin, a new 4U, 8/4 node server that joins our SuperBlade, GPU Platforms, SuperWorkstation, 4-Way and Twin architecture solutions for HPC.

insideHPC: Supermicro has been a platinum sponsor of ISC for years. What is
is about this event that keeps you coming back?

David Okada: Supermicro thrives on creating the best high-performance, high-efficiency platforms for supercomputing. It is our passion to perfect our solutions and offer this exacting quality to our customers so they can build supercomputing solutions that allow them to achieve their goals faster in the most cost-effective manner. Supermicro products are being increasingly selected by influential supercomputing customers so ISC is the perfect venue for us to show our leading-edge advanced supercomputing technologies to the customers who can deploy them.

Also posted in Events, HPC | Leave a comment

Video: HPC Market Update from IDC at ISC’12

In this video, Earl Joseph and Steve Conway from IDC present an HPC Market Update fro June 2012. Recorded at ISC’12 in Hamburg.

In related news, IDC will host two HPC User Forum meetings in Europe in July. The first meeting is at Imperial College London on July 5-6 and the second meeting is at the HLRS/University of Stuttgart on July 9-10Registration is free and refreshments will be served.

Also posted in Business of HPC, Events, HPC | Leave a comment

Part 2: ISC’12 News Blizzard

At insideHPC, we don’t seem much point in Xeroxing the myriad of press releases coming out of ISC’12. To bring you “HPC News Without the Noise,” here is a summary from Day 2 of the International Supercomputing Conference.

  • Eurotech announced an agreement with Nvidia to expand the company’s Aurora supercomputer product line with new energy-efficient, high-performance GPU-accelerated systems. The new systems are expected to deliver more than 500 teraflops of performance per rack and above 3.6 GFlops per watt.
  • Gnodal announced that the Rutherford Appleton Laboratory is using their GS-Series switches to implement a virtualized computational system for global climate research.
  • OpenACC is quickly becoming one of the industry’s preferred parallel programming solutions for accelerators.  While Directives-based compilers have been available for Nvidia GPUs for a while now, support in the works for platforms from AMD and Intel.
  • PRACE in Europe has introduced alternative HPC architecture prototypes focused on Resilience, Node Accelerators, and Low-Power. While some results from this effort are directly applicable to current systems, others will affect the next generation of supercomputers.
  • Terascala announced a new software platform for easily deploying, managing, and optimizing high throughput storage solutions. Already, Dell and EMC are integrating the software into their storage products.
  • Whamcloud has released version 1.0 of their Chroma Enterprise management software for Lustre. Both DataDirect Networks and NetApp have already announced they are building Chroma Enterprise-based appliances.


Also posted in Business of HPC, Events, HPC | Leave a comment

Advertisement

ClusterStor Ad

Video Archive

insideHPC.com is a production of insideHPC, LLC. © 2006-2013 Sitemap