Dell And Intel To Open Xeon Phi Centres Of Competence
Researchers will be able to get to grips with parallel computing ahead of Xeon Phi release
On Thursday, Dell and Intel announced plans to open two Product Centres of Competence (PCC) at the University of Cambridge, to serve as testbeds for Intel’s Xeon and Xeon Phi processors.
The Centres will offer High Performance Computing (HPC) training and the latest hardware to the European scientific research community, and are expected to open their doors before the end of the year.
Supercomputer wonderland
The PCCs will promote Intel’s Many Integrated Core (MIC) supercomputer architecture and Xeon Phi (earlier known as Knights Corner), a coprocessor which provides one teraflop of performance trough a card occupying just one PCIe slot.
The technology, announced in June, has already been dubbed a “Supercomputer on a Chip”, and is expected to go into production later this year.
Intel Xeon has been a staple of workstation, server, and embedded system markets since the late nineties. Xeon Phi coprocessors use the same well established programming language and tools, but excel at “highly parallel applications” – specialised tasks which require complex mathematical calculations.
In order to drive quicker adoption of Xeon Phi, the two Intel PCCs will enable researchers to learn, optimize and test their code using Intel Xeon and future MIC products.
The aim is to prepare the scientific community for the launch of the first generation of the Xeon Phi family of products, so this coprocessor can be used immediately as a production tool.
This initiative is expected to last for at least two years, and help improve combined threading and vectorisation – some of the biggest challenges in parallel applications. Dell says that while relatively few applications today are highly parallel, in the future they could address a wide range of important issues “ranging from climate change simulations and genetic analysis, to investment portfolio risk management and the search for new sources of energy”.
“Collaboration with the scientific research community is the foundation of Dell’s suite of high performance and high throughput solutions,” said Tim Carroll, global director of Dell’s Research Solutions Group.
“We provide solutions to scientific researchers whose needs range from workstation to Petascale class machines – exclusively on standards-based and open architectures. We are excited to bring this resource to life with our new PowerEdge C8000 series to drive innovation and discovery,” he added.
Dell and Intel are expected to announce one more PCC at a later date.
This year, Intel has been extremely active on the HPC market. Earlier this month, it started work alongside Hewlett-Packard on a highly energy-efficient supercomputer based around Xeon Phi. Installation will begin in November, with full capacity coming online in the summer of 2013.
In April, the chip manufacturer bought $140 million (£88m) worth of assets from supercomputer legend Cray, and in June the University of Cambridge and Imperial College London had launched CORE – the fastest Intel-based HPC system in the UK.
How much do you know about storage devices? Take our quiz!