ARCHIVE
NAG Delivers Numerical Software to Xeon Phi
The Numerical Algorithms Group (NAG) announces numerical software and services to support customer adoption of Intel Xeon Phi coprocessors. Routines from the NAG Library, running on the Intel Xeon Phi coprocessor, offload intensive computations to the coprocessor when it is efficient to do so. This enables users to exploit the power of the Intel Xeon Phi coprocessor transparently. You can see the first routines in action on booth #2431 at SC12 this week.
NAG will also provide software engineering and performance services for Intel Xeon Phi coprocessors, supporting customers migrate their own application codes to take advantage of Intel’s new product.
“The Numerical Algorithms Group has long provided value to the Intel Xeon processor with high quality optimized libraries for numerical computing,” said Joe Curley, Director of Marketing for Technical Computing Group at Intel Corporation. “In an extension of this partnership, NAG collaborated to provide input that helped in the definition and development of Intel Xeon Phi coprocessor technology. Tailoring the NAG Library to support Intel Xeon Phi coprocessors should benefit our mutual developers and customers’ highly parallel application development efforts.”
The NAG Library, now comprising over 1,700 routines, continues to provide an industry benchmark for accuracy, quality, reliability and documentation standards. Performance sensitive routines are updated with each library release to work efficiently on parallel computing systems using standards such as OpenMP and MPI. NAG has tuned and tested parallel routines for running on the Intel Xeon Phi coprocessor, which offers an order of magnitude more threads than traditional multicore processors.
As well as producing the NAG Library and providing HPC services, NAG also works with leading technology manufacturers by developing robust customised versions of complex numerical algorithms, by training their software engineers in parallel software engineering for numerical code and by participating in collaborative numerical library projects.
