In this issue:
NAG C Library algorithms have been engineered to execute efficiently on Cavium Inc.'s ThunderX ARMv8-A based Workload Optimized Processors. Preliminary results show excellent scaling across 96 cores of ThunderX in a dual socket configuration.
"NAG continues to demonstrate their advanced software expertise and leadership with this most recent delivery of optimized libraries for ThunderX," said Larry Wikelius, Vice President, Software Ecosystems and Solutions, Cavium. "NAG C Library algorithms will enable applications to take full advantage of scaling across all 96 cores of ThunderX, Cavium's ARMv8-A based Workload Optimized Processor with dual-socket support. NAG's team is recognized for their track record in engineering both high performing and accurate algorithmic results and Cavium is excited to see this partnership continue to provide critical software to the HPC market."
Learn more about the NAG C Library and the benefits of using it here. Collaboration is at the core of NAG's mission. If you're interested in forming a new collaboration, do contact us - we'd be delighted to hear from you.
In our latest Technical Report we examine the main theoretical aspects in some models used in Portfolio credit risk. We introduce the well-known Vasicek model, the large homogeneous portfolios or Vasicek distribution and their corresponding generalizations. An illustrative example considering factors following a logistic distribution is presented. Numerical experiments for several homogeneous portfolios are performed in order to compare these methods. Finally, we use the NAG Toolbox for MATLAB® for implementing prototypes of these models quickly.
Mathematical Optimization, also known as Mathematical Programming, is an aid for decision making utilized on a grand scale across all industries. Advanced analytical techniques are used to find the best value of the inputs from a given set which is specified by physical limits of the problem and user's restrictions. The quality of the result is measured by a user metric provided as a scalar function of the inputs. Optimization problems come from a massively diverse range of fields and industries, such as portfolio optimization or calibration in finance, structural optimization in engineering, data fitting in weather forecasting, parameter estimation in chemistry and many more.
NAG's Mathematical Optimization Consultancy experts provide you with all the information needed to solve an optimization problem. They will discuss your current approach and help guide you to use the right solver for the particular problem. If you need to improve the performance, we can assess how your current model fits the particular solver and advise on a possible reformulation or tune the solver for you. If the problem requires it, NAG's software experts can adapt an existing solver or develop an entirely new one.
If you have a problem that could benefit from expert analysis and advice, contact us.
Within the NAG Library is a chapter of optimization routines that are expertly developed, documented, supported and maintained. For more information on these specific NAG Library solvers click here.
NAG's optimization team helped an Oil & Gas client, PSI AG, formulate their gas pipeline problem as a mixed integer linear program (MILP). NAG offered a suitable solver and a modelling environment to efficiently solve the problem and its variants.
The new solution provides substantial improvements and benefits. The complete optimization problem has been solved within 0.5 seconds.
"The help that NAG services provided was invaluable for us," said Michael Krätsch of PSI. "The increase in speed that we are now able to achieve, when solving our specific pipeline problem, was amazing and the knowledge we were given about optimization solvers, problem formulation and modelling techniques was more than we hoped for. The NAG people are a pleasure to work with and I couldn't have wished for more from any consultancy engagement."
Learn more about how NAG experts helped PSI here.
Global Data and the NAG Toolbox for MATLAB®
Some of the most useful and powerful routines in the NAG Toolbox for MATLAB® require the user to provide a function, supplied as either an M-file or as a function handle. For example optimization routines require the user to provide the function to be minimized. Ordinary differential equation routines and quadrature routines also require the user to provide functions in order to specify the problem. Read the blog here.
Numerical Optimization and Scaling
Scaling can often have a significant influence on the performance of an optimization routine. Currently there are no user-callable scaling routines in the NAG Libraries, but scaling can be performed automatically in routines which solve Sparse Linear Programming, Quadratic Programming (QP) or Nonlinear Programming (NLP) problems and in some dense solve routines. Read the blog here.
Finding the Right Routine in the NAG Library - Decision Trees
One of the great features of the NAG Library is the sheer number of routines: over 1,700 in total. However, this can make the task of identifying the right routine a little daunting and time consuming. In this blog post we highlight the decision trees included in each of the NAG Library Chapter Introductions found in the accompanying Library documentation. Read the blog here.
Come and see us at various conferences and events over the next few months.
- Fortran Modernisation Workshop
28-29 July 2016, The University of Southampton
- Advanced Risk and Portfolio Management (ARPM) Bootcamp by Attilio Meucci
15-20 August 2016, New York University
New York University (40 GARP CPD and 40 CFA Institute CE credits and ARPM Certificate®) To register with the Discounted Affiliate Rate go to: Registration, then 1) select Affiliate; 2) click on Next Step; and 3) specify "NAG" in the ARPM-affiliated entity field. Or contact firstname.lastname@example.org.
- Introduction to Modern Fortran Course, delivered by FortranPlus
- Fortran Modernization Workshop at Culham Centre for Fusion Energy
24-25 August 2016, Oxfordshire
- Fortran Modernization Workshop at Queen Mary University London
1-2 September 2016, London
- GPU Technology Conference
28-29 September 2016, Amsterdam