Adapteva Kickstarts credit card-sized supercomputer project

October 03, 2012

Bernard Cole-October 03, 2012

Lexington, Mass. - Adapteva, a semiconductor start-up company has just initiated the crowd-funded Parallella Kickstarter project  to build an open and affordable $99 credit card-sized supercomputing platform that offers a 50x performance boost over existing open hardware platforms like Raspberry Pi.

The disruptive performance leap of the Parallella computer  is enabled by Adapteva’s 16-core and 64-core Epiphany microprocessor chips.

The present and future of computing is parallel, said Andreas Olofsson, CEO of Adapteva, but developers are facing enormous challenges in developing code that can take advantage of the exponentially increasing performance offered by today’s parallel architectures.

Without wide-spread adoption of efficient parallel programming methods by the developer community, he said, application performance improvements will soon hit the wall.

“Inspired by great hardware communities like Raspberry Pi and Arduino, we see a critical need for a truly open, high-performance computing platform that will close the knowledge gap in parallel programming,” said Andreas Olofsson, CEO of Adapteva. “High-performance parallel computing is currently only accessible to a small and select group of expert programmers. The goal of the Parallella project is to democratize access to parallel computing. The parallel programming dilemma is a grand challenge of computer science and we need all hands on deck.”

Numerous universities seeking to educate students on parallel computing already are collaborating with Adapteva. To date, Adapteva has collaborated with researchers at MIT, Boston University, Northeastern University and Halmstad University to help bring parallel computing technology to the next generation of engineers.

“Processor architectures need to be built on simple, scalable principles with powerful processors and straightforward means for moving data around,” said Bertil Svensson, Professor at Halmstad University. “My students and researchers have had a positive experience with Epiphany, using both the architectural features and programming tools efficiently on their first try.”

To create a sustainable community around the Parallella computer platform, Adapteva will be raising a minimum of $750,000. The funds will enable Adapteva to bring costs down for its existing evaluation boards and deliver a fully functional computer that includes a dual core ARM A9 based SOC and a 16-core Epiphany-III co-processor chip for $99.

Olofsson said the 16-core board Parallella should achieve 13 GHz and 26 GFLOPS of equivalent CPU performance. If the group-funded Parallella Kickstarter project reaches a stretch goal of $3M, a higher performance Parallella board will be released that includes the pin-compatible 64-core Epiphany-IV chip.

He thinks the Epiphany-IV based board should achieve a peak performance of 45 GHz and 90 GFLOPS and will be available to all backers who pledge $199 or more. The Parallella computers will ship with a Linux Ubuntu distribution and an open source SDK for developing applications for the Epiphany architecture using C, C++ and/or OpenCL.

Once the Parallella project is fully funded, Adapteva will release its existing software development tools, drivers and libraries under a true open source license and will publish the Epiphany chip architecture reference manuals and datasheets publicly.

Board design files and board support packages for the Parallella computer will be delivered to the public in open source format and available for everyone to use free of charge and without restrictions.

“I truly believe that the availability of open and inexpensive hardware, free open source software development tools, and increasingly knowledgeable parallel programmers is what will drive the next wave of innovation in software applications,” said Olofsson. “If we as a community can offer the right building blocks, programmers will change the world in ways we have yet to imagine.”

For more information, go to www.adapteva.com.

Loading comments...

Most Commented

  • Currently no items

Parts Search Datasheets.com

KNOWLEDGE CENTER