High-Performance Computing | News
Intel To Target 'Many-Core' Chips at Highly Parallel Applications
- By Dian Schaffhauser
Intel will soon go into production with a Many Integrated Core (MIC) architecture that uses the x86 instruction set to create high-performance computing platforms running at trillions of calculations per second. The company made the announcement during an international supercomputing conference.
The first product in the new line, code-named "Knights Corner," will be made with Intel's 22-nanometer manufacturing process. This process produces transistor structures as small as 22 billionths of a meter and will place 50 processing cores on a single chip.
The new line of chips is intended for use in highly parallel applications, targeting segments such as exploration, scientific research, and financial or climate simulation.
Intel has begun making design and development kits, codenamed "Knights Ferry," available to a group of developers and said it will expand the program during the second half of 2010.
The MIC architecture is derived from several Intel projects, including "Larrabee," a many-core x86 architecture for graphics cards (a project the company has put on hold); and the "Single-chip Cloud Computer," which researches how to accelerate futuristic chip development with 10 to 20 times the processing power of current consumer processors.
"The CERN openlab team was able to migrate a complex C++ parallel benchmark to the Intel MIC software development platform in just a few days," said Sverre Jarp, CTO of CERN openlab. "The familiar hardware programming model allowed us to get the software running much faster than expected."
Dian Schaffhauser is a senior contributing editor for 1105 Media's education publications THE Journal and Campus Technology. She can be reached at firstname.lastname@example.org or on Twitter @schaffhauser.