Entries filed under “Green HPC”

Design and management techniques that contribute to the responsible, effective use of energy in the operation of high performance computing centers and equipment.

New Technical Report: Energy Rooflines

Over at Georgia Tech’s HPC Garage, Rich Vuduc writes that a new technical report presents a thought-experiment on the question of whether engineering an algorithm to optimize time differs from doing so with respect to energy Joules.

Our goal is to explain—in simple, analytic terms accessible to algorithm designers and performance tuners—how the time, energy, and power to execute an algorithm relate. The model considers an algorithm in terms of operations, concurrency, and memory traffic; and a machine in terms of the time and energy costs per operation or per word of communica- tion. We confirm the basic form of the model experimentally. From this model, we suggest under what conditions we ought to expect an algorithmic time-energy trade-off, and show how algorithm properties may help inform power management.

Download the Report (PDF).

Also posted in Computing Research, HPC | Leave a comment

Video: Interview on Energy & HPC with Jack Dongarra and Dona Crawford

In this video, Dan Olds from Gabriel Consulting sits down with Jack Dongarra (ORNL/University of Tennessee and Dona Crawford (Assoc Director LLNL) at SC12 to discuss the challenges facing HPC on the road to exascale. Along the way, they describe their TOP500-list-topping systems: Titan and Sequoia.

Also posted in Events, HPC, SC12, Video | Leave a comment

Infographic: Why Green Datacenters Matter

Over at the Intel CSR blog, Lorie Wigle writes that replacing an old server with a more energy efficient design could save up to 1 ton of carbon emissions.

Click image to expand.

Intel is also exploring other innovative and eco conscious technologies within its data centers. For example, on its Rio Rancho, New Mexico campus, Intel conducted a pilot program with mineral oil cooling. For an entire year, the Rio Rancho servers were completely submerged in vats of mineral oil, which removed any excess heat and in turn, improved cooling for the entire data center. At first thought, the idea of placing technology equipment in mineral oil sounded pretty strange, but the results quickly changed my mind: The mineral oil cooling allowed us to save 7 percent of the server power by removing the fans, and modeling suggests it could result in a 90 to 95 percent reduction in energy use for the overall data center cooling system.

Read the Full Story.

Also posted in Datacenter operations, HPC | 3 Comments

Lowering Trends in Energy per Operation

Over at Datacenter Knowledge, Intel’s Winston Saunders looks at SPECpower data and how extended efforts to increase the efficiency of servers under realistic workload scenarios has resulted in a 40% per year reduction in the energy per operation.

So to summarize, the efficiency of two-socket servers has increased dramatically. The efficiency gains result from the increased performance and energy proportionality of the systems. Transitions in transistor architecture result in large performance gains resulting in a reduction in the time necessary to complete a given workload. Transitions in system architecture also result in performance gains, but in addition lower the power consumption while the operation is being completed. This in some sense “doubles” the efficiency gains expected. This efficiency gain is easily visualized by looking at the time to complete an operation and the power used during that time.

Read the Full Story.


Also posted in Compute, HPC, HPC Hardware | Leave a comment

European Mont-Blanc Project Selects Samsung Multicore ARM Processor

The Mont-Blanc European project has selected the Samsung Exynos platform as the building block for powering its first integrated low power-high performance computing (HPC) prototype.

The aim of Mont-Blanc project is to design a new type of computer architecture capable of setting future global HPC standards, built from today’s energy efficient solutions used in embedded and mobile devices.

The Samsung Exynos 5 Dual is built on 32nm low-power HKMG (High-K Metal Gate), and features a dual-core 1.7GHz mobile CPU built on ARM Cortex-A15 architecture plus an integrated ARM Mali-T604 GPU for increased performance density and energy efficiency. It has been featured and market proven in consumer and mobile devices such as Samsung Chromebook and Google’s Nexus 10.

This will be the first use of an embedded mobile SoC in HPC, which enables the Mont-Blanc project to explore the challenges and benefits of deeply integrated energy-efficient processors and GPU accelerators, compared to traditional homogeneous multicore systems, and heterogeneous CPU + external GPU architectures.

The Exynos 5 Dual packs the most powerful ARM processors with a programmable GPU in a low-power mobile device that would normally be in someone’s pocket and running on a battery. Its performance density, energy efficiency, and low market price make it an extraordinary building block for prototyping a new generation of HPC systems.’ said Alex Ramirez, coordinator of the Mont-Blanc project.

During the first year of activities, Mont-Blanc has focused on deploying successfully an HPC system software stack and full-scale scientific applications on ARM platforms, proving that ARM-based architectures are feasible alternatives for HPC. Now the efforts gear towards integration of the Exynos platform on a HPC solution, and software exploitation of the embedded GPU.

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

Also posted in Compute, Computing Research, HPC, HPC Hardware | Leave a comment

Texas Instruments Offers System on a Chip for HPC Applications

In this video from SC12, Arnon Friedmann from Texas Instruments describes the company’s new multicore System-on-Chips (SoCs). Based on its award winning KeyStone architecture, TI’s SoCs are designed to revitalize cloud computing, inject new verve and excitement into pivotal infrastructure systems and, despite their feature rich specifications and superior performance, actually reduce energy consumption.

Using multicore DSPs in a cloud environment enables significant performance and operational advantages with accelerated compute intensive cloud applications,” said Rob Sherrard, VP of Service Delivery, Nimbix. “When selecting DSP technology for our accelerated cloud compute environment, TI’s KeyStone multicore SoCs were the obvious choice. TI’s multicore software enables easy integration for a variety of high performance cloud workloads like video, imaging, analytics and computing and we look forward to working with TI to help bring significant OPEX savings to high performance compute users.”

Read the Full Story.

In related news, TI announced today that Nimbix will use the company’s high-performance KeyStone multicore DSPs, significantly reducing power and accelerating workflows for video processing and imaging applications in the cloud.

Also posted in Accelerators, Events, HPC, HPC Hardware, SC12 | Leave a comment

Video: Argonne to Lead Energy Storage Hub for 5X Better Batteries

Our Video Sunday feature continues with this press conference by U.S. Secretary of Energy Steven Chu announcing that Argonne National Laboratory has been selected for an award of up to $120 million over five years to establish a new Batteries and Energy Storage Hub. The Hub, to be known as the Joint Center for Energy Storage Research, will combine the R&D firepower of five DOE national laboratories, five universities, and four private firms in an effort aimed at achieving revolutionary advances in battery performance.

The goal of this effort is to deliver battery technology with 5X more capacity and efficiency in five years. Imagine a Nissan Leaf vehicle with a range of 500 miles on a charge or a solar-powered battery backup for your home for less than $10K.

This is a partnership between world leading scientists and world leading companies, committed to ensuring that the advanced battery technologies the world needs will be invented and built right here in America,” said Secretary Chu. “Based on the tremendous advances that have been made in the past few years, there are very good reasons to believe that advanced battery technologies can and will play an increasingly valuable role in strengthening America’s energy and economic security by reducing our oil dependence, upgrading our aging power grid, and allowing us to take greater advantage of intermittent energy sources like wind and solar.”

Advancing next generation battery and energy storage technologies for electric and hybrid cars and the electricity grid are a critical part of President Obama’s all-of-the-above energy strategy to reduce America’s reliance on foreign oil and lower energy costs for U.S. consumers. And HPC will make it possible. Amazing. Read the Full Story.

Also posted in Computing Research, HPC, National and Legislative Action, Video, Video Sunday | Leave a comment

The Green Grid Offers New Energy-saving Resources for Datacenter Managers

Over at The Green Grid, Director Roger Tipley writes that a major challenge for his organization is helping datacenter managers to recognize how far they should be pushing the efficiency agenda in their organizations.

If your car used virtually the same amount of energy when traveling at 80 miles per hour as when it was parked, you would assume there was something seriously wrong with the engine. But this is essentially how many data centres function. There have been many efficiency improvements stemming from areas such as cooling and virtualisation. However, the great challenge remains to minimise the energy consumed when parts of the servers are idle. We need to get IT leaders to push harder in this front of the war on DC efficiency.”

Along these lines, The Green Grid offers these new resources:

Read the Full Story.

Also posted in HPC | Leave a comment

Asetek Reinvents the Datacenter with Innovative Hot Water Cooling at SC12

In this video from SC12, Asetek founder & CEO André Eriksen describes the company’s innovated hot water cooling technologies for HPC and Cloud computing.

ISAC is revolutionary, completely eliminating the need for CRAC (Computer Room Air Conditioning) in the data center. All air inside the server stays inside the server and recirculates rather than exiting and heating up the data center which also significantly relaxes the air dust quality spec inside the data center. Each CPU is liquid cooled with Asetek’s proprietary liquid cooling while a liquid-to-air heat exchanger inside the server cools the inside server air. Each component inside the server sees the same temperature and air-flow as it would in a traditional data center install.”


Also posted in Events, HPC, SC12, Video | Leave a comment

Intel Completes Testing of CarnotJet Submersion Cooling

Intel has endorsed Green Revolution Cooling’s CarnotJet submersion cooling system following one year of testing. Intel found that submerged servers used 7 per cent less energy at platform level due to removal of server fans and reduction of leakage currents in mineral oil coolant, and reported a cooling PUE (Power Usage Effectiveness) of 1.02-1.03, which means the CarnotJet system requires just three per cent of IT power to cool the IT servers. Modern data centres typically have a PUE of 1.6-1.7 and air-cooled servers that use 5-25 per cent more energy than submerged servers.

According to Green Revolution Cooling (GRC), Intel’s support for submersion cooling has bolstered sales of the product and the company has announced an installation base of more than a megawatt with a number of new installations on the way. GRC has also recently hired a Systems and Controls team to further refine the control system I/O for data centre integration. In addition, a factory server warranty from Supermicro for submersion-ready servers installed in the CarnotJet system has been announced.

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


Also posted in Computing Research, HPC | Leave a comment

Beacon System Rises to Top of Green500 with 2.5 Gigaflops per Watt

Over at Forward Thinking, Michael J. Miller writes that the top system on the Green500 achieved an amazing 2.5 gigaflops per watt.

This time, the top system is a relatively small cluster (#253 on the TOP500 list), the Beacon system run by the National Institute of Computational Science and the University of Tennessee. It is built using Appro’s GreenBlades, each running 2.6GHz Xeon E5-2670 processors (8 cores/16 threads) and Xeon Phi 5110P co-processors, using Infiniband FDR interconnects.

Read the Full Story.


Also posted in HPC | Leave a comment

Slidecast: New Texas Instruments SoC’s Reduce Power Consumption

In this slidecast, Arnon Friedmann from Texas Instruments describes the company’s new multicore System-on-Chips (SoCs). Based on its award winning KeyStone architecture, TI’s SoCs are designed as accelerate traditional x86 servers as well as enable the building purpose-built, energy efficient devices for specific applications powered by ARM processors and TI DSPs on the same package.

Targeted for applications such as networking, radar, imaging, high performance computing, gaming and media processing, TI’s new KeyStone multicore processors offer developers more than twice the capacity & performance at half the power relative to existing solutions.

Read the Full Story * Download the MP3Subscribe on iTunes * If Dropbox is blocked, Download from Google Drive.

Also posted in Accelerators, Events, HPC, HPC Hardware, Podcast, Rich Report, SC12, Video | Leave a comment

Saudi Supercomputer is Tops in Energy Efficiency

This week the Saudi supercomputer “SANAM” was ranked #2 Green500 listing of the most energy-efficient computers. Developed as part of the joint project of the leading Saudi KACST research organization and the Frankfurt Institute for Advanced Studies, the system also ranks #52 on the TOP500 and delivers an outstanding 2,351 million calculations per second per watt.

SANAM will be utilized to support research activities by many institutions in Saudi Arabia,” said Dr. Turki Alsaud, KACST VP for research institutes. “In this project, I’m pleased that we had a team of five engineers and two PhDs from KACST worked jointly with our German partner to develop SANAM.”

The Arabic name “SANAM” reflects performance and efficiency. Comprising 210 ASUS servers and 3360 processor cores, SANAM is a hybrid system with Intel Xeon E5-2650 processors and AMD FirePro S10000 graphic cards, all connected with FDR InfiniBand. Read the Full Story.

In related news, insideHPC is headed to the Saudi Arabian High Performance Computing Conference, which will be held December 1-3, 2012 at KAUST University. The conference is designed to nurture HPC communities in the region, facilitate information exchange, foster new collaborations, and advance the development of related applications and scholarships.

Also posted in Events, HPC, HPC Saudi Arabia, New Installations | Leave a comment

The Green Grid Partners with Global Task Force on Energy Metric Consensus

This week The Green Grid global task force announced three new efficiency metric measurement guidelines: the use of the Green Energy Coefficient (GEC), Energy Reuse Factor (ERF) and Carbon Usage Effectiveness (CUE) metrics.

The guidelines include:

  • Green Energy CoefficientGEC quantifies the portion of a facility’s energy that comes from green sources. The metric is computed as green energy consumed by the data center (kWh) divided by total energy consumed by the data center (kWh).
  • Energy Reuse FactorERF identifies the portion of energy that is exported for reuse outside of the data center. ERF is computed as reuse energy divided by total energy consumed by the data center.
  • Carbon Usage Effectiveness CUE enables an assessment of the total GHG emissions of a data center relative to its IT energy consumption. CUE is computed as the total carbon dioxide emission equivalents (CO2eq) from the energy consumption of the facility divided by the total IT energy consumption.

We want to make it as easy as possible for the data center community to understand and embrace these metrics, and how they can work together,” said Joyce Dickerson, Board Member, The Green Grid. “When implemented correctly, they can save organizations a lot of time, money, and additional resources. Although there is more work to do, we think this will bring us one step closer to a universally adopted set of metrics, indices, and measurement protocols that will have a positive impact on the industry.”

The Green Grid and the task force intends to continue collaboration in the ongoing effort to improve data center energy and Greenhouse Gas (GHG) emission efficiencies. Read the Full Story.

Also posted in HPC | Leave a comment

TI’s New DSPs Increase Performance and Reduce Power Consumption

This week Texas Instruments (TI) unveiled a better way to cloud with six new multicore System-on-Chips (SoCs). Based on its award winning KeyStone architecture, TI’s SoCs are designed to revitalize cloud computing, inject new verve and excitement into pivotal infrastructure systems and, despite their feature rich specifications and superior performance, actually reduce energy consumption.

Using multicore DSPs in a cloud environment enables significant performance and operational advantages with accelerated compute intensive cloud applications,” said Rob Sherrard, VP of Service Delivery, Nimbix. “When selecting DSP technology for our accelerated cloud compute environment, TI’s KeyStone multicore SoCs were the obvious choice. TI’s multicore software enables easy integration for a variety of high performance cloud workloads like video, imaging, analytics and computing and we look forward to working with TI to help bring significant OPEX savings to high performance compute users.”

Stop by the Texas Instruments booth #4828 at SC12. Read the Full Story.

Also posted in Events, HPC, HPC Hardware, SC12 | Leave a comment

Podcast Series

Green HPC Podcast

Video Archive

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