UK Scientists Choose Cray for HECToR

By Christopher Lazou

March 30, 2007

The UK's Engineering and Physical Sciences Research Council (EPSRC) completed the procurement of a new large-scale computer system named 'HECToR' when it signed two contracts on 15 February 2007, each lasting 6 years. Cray Inc. in partnership with the University of Edinburgh-led HPCx Ltd is to provide the technical solution (hardware) and service operation; while the computational science and engineering (CSE) support service for academic users is to be provided by NAG Ltd. The process for identifying the requirements for HECToR took about three years and consisted of several phases involving the scientific community, such as defining user requirements, developing benchmarks and finally going to open tender.

As part of a series of reviews, there was an international evaluation of research using HPC in the UK. The international panel was chaired by Dr Horst Simon from NERSC and included eminent scientists from Europe, the USA and Japan. They published their findings in December 2005 and these were their recommendations:

1. Create a more balanced HPC infrastructure between computational technologies and intellectual resources.

2. Strengthen the computational infrastructure at universities by a) systematically deploying leading-edge capability systems, large-scale capacity computing systems and other resources; b) supporting and developing a state-of-the-art applications software infrastructure encompassing algorithms, data management and analysis, visualization, and best practices software engineering.

3. Develop human resources in HPC as the future of CSE depends critically on the availability of highly specialised experts. Because of its interdisciplinary nature, research using HPC requires specialists who have received additional education and training beyond traditional academic disciplinary programmes.

4. Bridge disciplines and build a computational science community by increasing interactions between disciplinary groups nationally and internationally.

The panel noted that: “while there are several examples of excellent collaborations in individual disciplines, an all embracing computational science community (in the sense of an 'academic community') does not yet exist in the UK. The panel recommends taking proactive steps towards the creation of such a computational science community….”

As the panel states in their executive summary: “Computational science, the scientific investigation of physical processes through modelling and simulation on computers, has become generally accepted as the third pillar of science, complementing and extending theory and experimentation”.

The writers of this report were right to claim that this view of computational science as the third pillar of science probably evolved in the mid-1980s, as those of us involved in supercomputing used it as our mantra at the time. It grew out of an impressive list of accomplishments in such diverse areas as astrophysics, aeronautics, chemistry, climate modelling, combustion, cosmology, earthquake prediction, imaging, materials, neuroscience, oil exploration, and weather forecasting. These accomplishments were achieved thanks to the Cray vector system of the time, which enabled the scientific community to perform meaningful modelling in their scientific field.

Back to the future. The HECToR system will be housed at the University of Edinburgh's Advanced Computing Facility (ACF). Because of the need for building modifications prior to the installation of this system, the HECToR service is expected to start in October 2007, with an 'Early Bird' service for selected users commencing in September 2007.

A one teraflop system for early test and development is to be delivered in May 2007 and will be used for porting codes. The main system comes in several phases over the next three years.

Initial installation will be a 60 teraflop Cray XT4 MPP later this year, followed by successive upgrades that will take the installation to approximately 270 teraflops in the late 2009 timeframe. The upgraded system will be a hybrid supercomputer that combines Cray's scalar MPP and vector architectures.

It may have surprised some that Cray won HECToR, the largest European HPC tender, but this contract award confirms that the HPC community is buying into the Cray vision of adaptive supercomputing. This procurement also highlights the capability prowess of the Cray system on offer, with its balanced, tightly integrated scalar-vector architecture that is the centrepiece of the Cray adaptive supercomputing vision. It is a true capability system, as befits a “capability-type” investment.

As I understand it, although I haven't seen the detailed report, Cray won based on a number of complex criteria, not only performance (and when I say performance I mean sustained performance for real applications, not peak or Linpack results). In addition, Cray was able to demonstrate scaling of some of the benchmarks to over five thousand cores. Note also that the Cray proposal for phase II of HECToR contains future (2009) technology that is also included in the Cray DARPA proposal.

The HECToR contract also includes two other components, a Cray support model with 3-4 people located onsite at the University of Edinburgh and a Cray centre of excellence (CoE).

As part of HECToR, Cray committed to establish a centre of excellence, which will directly link into Cray Research & Development and will last for the entire duration of the project. The Cray HECToR CoE joins three other such centres located at ORNL, KMA and NERSC. These centres will all interoperate to:

  • Accelerate the pace of innovation in supercomputing technology for technologies and applications that are relevant for the users;
  • Maximise usage of the computational resources;
  • Train and educate the next generation of interdisciplinary computational researchers to enhance the science delivery;
  • Provide a direct interface to Research and Development in Cray for any kind of engineering design change request or any other topic that may need the attention of the Engineering Division of Cray; and
  • Foster a climate of information exchange and collaboration between Cray HPC users.

The members of the Cray CoE will directly engage with scientific users of HECToR and will work closely with CSE (NAG Ltd.) and HPCx. This CoE is in addition to the on-site Cray support staff for the Cray supercomputer system. The CoE sits comfortably with the findings of the international review panel described above.

Like its predecessors, the CSAR and HPCx, systems purchased by EPSRC, HECToR is the flagship capability system for UK academic research.

The CSE support service for HECToR, to be provided by NAG Ltd, will operate differently from previous services on its predecessor systems such as CSAR and HPCx. A small central 'core' CSE support team will provide a limited level of assistance to users. In depth CSE support will be provided by a 'distributed' CSE support team. This distributed CSE support team will be based locally within the user research consortium, as agreed by NAG Ltd and the grant holder. Calls will be issued on a regular basis for research consortia to apply for distributed support of three months to up to two years.

The NAG service model is a textbook copy of the first recommendation of the International review of UK research report. It states: “…HPC platforms generally can be utilized more effectively if users have access to intellectual resources, for example experts in code optimisation, parallel algorithms, etc., who can work with the researchers as consultants or collaborators. A relatively small group of such experts, for example associated with HECToR, can achieve high productivity gains if they are leveraged across multiple users. The productivity gains are obtained by more rapid insertion of new technology into applications codes and by realization of technology transfer between research groups”.

Access to the HECToR service is allocated by requesting time on research proposals and is open to all EPSRC and Biotechnology and Biological Sciences Research Council (BBSRC) eligible researchers. These applications will be accepted for consideration from May 2007. Full details of the application process will be available on the EPSRC Website by May 2007.

It is expected that a different process will be set up for the Natural Environment Research Council (NERC) supported researchers, as the process used by NERC to allocate time on HECToR differs from that of EPSRC and BBSRC.

HECToR, one of the largest and most ambitious high performance computing (HPC) initiatives in Europe, costing £113 million over six years, is funded largely from the UK large facilities capital fund, established to ensure UK scientists have access to leading edge, large-scale experimental projects and facilities. EPSRC, NERC and BBSRC are also contributing financially, emphasising the breadth of science that will be supported.

As innovation in information technology is gathering pace, opportunities for computational science multiply. HECToR is the latest acquisition in this continually evolving area of HPC. When HECToR comes on stream, scientists from all areas of the UK science base will have the best possible computer facilities, enabling them to carry out cutting edge research and remain at the forefront of research internationally.

UK science has a long history of collaboration. This philosophy was enshrined in the Collaborative Computational Projects (CCPs) enacted in the late 1970s, and it persists to this day in parallel with the newer consortium approach adopted in the 1990s. Taking a glance at the 2006 issue of CSE Frontiers, the annual report of the Council for the Central Laboratory of the Research Councils (CCLRC) CSE department, one can see what tremendously important work is produced under these collaborative schemes. Be it in astrophysical and laser-produced plasmas, under CCP2; helicopter rotor wakes simulations, electronic structure of semiconductor quantum dots, materials chemistry, superconductivity under extreme pressure, valence transitions in nickelates, and state-of-the art software for experimental determination of protein structures, under CCP4; or nano-objects in supercritical media and molecular dynamics study of radiation hard materials, e.g. radiation damage in pyrochlores. The list is long, almost endless.

As an illustration, the Collaborative Computational Project No. 4 (CCP4) was set up in 1979 to coordinate the efforts of UK protein crystallographers who were writing software to help with the structure determination steps. Since then, the project has grown considerably, and now is the leading provider of PX software worldwide. The CCP4 suite includes around 200 programs covering all aspects of structure solution by protein crystallography. The programs are typically accessed via a graphical interface that controls project organization, job submission and display of results.

The purchase of HECToR chimes well with the European Forward Look initiative and also dovetails with the concepts enshrined in the 7th European R&D Framework Programme (FP7) where Europe plans to spend over 50 billion Euros in the next six years on research and development, including a strong element on HPC infrastructure. The recent announcement of setting up a European Institute of Technology (EIT) is but one example. EIT aims to fund 10 Knowledge and Innovation Communities (KICs), 4,000 to 5,000 scientists, 6,000 Master's students and 4,000 PhD candidates in the next eight years at an annual cost of 2 billion Euros to enhance new interdisciplinary educational training and develop human resources.

The KICs will operate across Europe and will cover such strategic topics as climate change, renewable energy, information society, health science and technology, nano-technologies, and so on. The KICs will be funded for between seven and fifteen years, giving them ample time to demonstrate their mettle.

There is also a provision in FP7 for Europe to acquire three to four systems in the petaflops range at the earliest possible time to be used by scientific researchers across Europe to boost innovation capability and improve European economic competitiveness.

In summary, for UK scientists, HECToR should enable them to maintain their position amongst the leaders in advanced computational science and engineering worldwide. Cray was selected by EPSRC for promising a capability system of scalable and balanced architecture with high-sustained performance, a centre of excellence, and in 2009, a scalar-vector, next generation MPP technology.

As Ulla Thiel, Cray Vice President for Europe, put it: “For Cray Inc. this sale highlights Cray's rapidly increasing presence in the European HPC community and will serve as an excellent showcase for Cray's most advanced computing concepts, bringing to bear the capabilities of Adaptive Supercomputing to address the demanding needs of leading-edge scientific research”.

—–

Copyright (c) Christopher Lazou, HiPerCom Consultants, Ltd., UK. March 2007. Brands and names are the property of their respective owners.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire