ISC13, Leipzig, Germany
17th June 2013 - The NAG Library for SMP & Multicore has been extended and now includes even more parallelized algorithms enabling users of shared memory systems to solve their numerical problems faster. Newly parallelized routines are applicable to problems in the areas of global optimization, matrix functions and statistics, including Gaussian mixture model, Brownian bridge and univariate inhomogeneous time series. This update to the NAG Library for SMP & Multicore will greatly benefit software developers wishing to easily exploit the performance potential of multicore systems without having to learn the intricacies of parallel programming, as well as those running large computations where execution speed is imperative.
The NAG Library for SMP & Multicore contains the same mathematical and statistical content as the standard NAG Fortran Library – over 1,700 routines – but over a third of the routines in the Library for SMP & Multicore can now utilize multiple cores. This feature enables applications to take advantage of the latest processor and memory technology and that means the routines run faster and more efficiently on multicore processors. No other commercially available numerical library has the breadth of numerical functionality that is found in the NAG Library for SMP & Multicore.
NAG has a longstanding reputation for providing access to reliable numerical analysis techniques and adapting them to be suitable for use on leading edge hardware. The NAG Library is used in fields that are as diverse as scientific modelling, weather forecasting, financial modelling and complex engineering design. NAG is also well-known to many leading technology organisations that base their products on NAG numerical code.
Speaking of the latest release of the NAG Library for SMP & Multicore, Dr Reinhold Bader of Leibniz Supercomputing Centre (LRZ) in Garching, Germany, said “The NAG Library for SMP & Multicore has been deployed on the flagship HPC systems at LRZ for more than two decades and we welcome the added functionality in the Mark 24 release. The superior scaling of the provided computational kernels in shared memory can provide a significant performance advantage to HPC applications that use hybrid parallelism. Furthermore, we intend to test a specially tuned version of this Library also on an upcoming Intel Xeon Phi installation later this year.”
The newly developed NAG Library routines are available with OpenMP support and will be showcased at ISC'13 in Leipzig 16-20 June 2013 on booth #650.
Contact at ISC’13: Marcin or Viktor on booth #650
Image shows examples of NAG Library for SMP & Multicore performance at Mark 24. Each line represents a different problem size. (Platform: AMD Opteron 6174 processors. Each core running at 2.2 GHz.)
The Numerical Algorithms Group (NAG) is dedicated to applying its unique expertise in numerical engineering to delivering high-quality computational software and high performance computing services. For over 40 years NAG experts have worked closely with world-leading researchers in academia and industry to create powerful, reliable and flexible software which today is relied on by tens of thousands of individual users, as well as numerous independent software vendors. NAG serves its customers from offices in Oxford, Manchester, Chicago, Tokyo and Taipei, through staff in France and Germany, as well as via a global network of distributors.