NAGNews 97

Welcome to NAGNews

Issue 97, 19 May 2011

In this issue


Calling the NAG Pseudo and Quasi Random Number Generators From a Multi-Threaded Environment


The NAG Library features a collection of Random Number Generators. In this new article, and the associated example programs, we will show how to call the NAG random number generators within a multi-threaded environment. The examples are written making use of OpenMP, however the basic structure of the NAG calls will be the same irrespective of the threading mechanism used. Some OpenMP commands and pragmas are briefly described in this document, however additional information is available on the OpenMP website, alternatively NAG offer training courses in OpenMP, details of which can be obtained by email.

To read the article click here


NAG Numerical Routines for GPUs updated


NAG have updated the NAG Numerical Routines for GPUs to include new functionality for developers of applications that utilize GPUs.

The routines available now are:

  • The Multiple Recursive Generator MRG32k3a described in L'Ecuyer (1999)
  • The Mersenne Twister MT19937 described in Matsumoto and Nishimura (1998)
    (Note: NAG allows skip aheads for the Mersenne Twister, so that independent streams and substreams can be constructed)
  • The Sobol generator is based on the work of Joe and Kuo (2003) and Joe and Kuo (2008) and yields a generator in 21,201 dimensions. The pseudorandom extension proposed by Jackel increases the capability to 50,000 dimensions.
  • Normal, Uniform and Exponential distribution generators
  • A modified depth-ordered algorithm for Brownian bridge

In addition, matching serial routines are now also provided to enable duplicate CPU based systems to be designed to help with development, test and compliance processes.

For more information click here.


NAG Student Prize Winning Paper - Toward Adjoint-Based Optimization in Computational Finance


NAG remains committed to supporting the numerical work of students all over the world. The most recent award - the NAG & ICBI Financial Mathematics Student Prize was awarded to Niloofar Safiran of RWTH Aachen University at the Global Derivatives & Risk Management conference. Niloofar's winning paper entitled "Toward Adjoint-Based Optimization in Computational Finance" can be viewed here. Congratulations Niloofar.

Niloofar Safiran
John Holden of NAG with Niloofar Safiran - RWTH Aachen University and Marie Houghton - ICBI

To learn more about NAG's work in the area of Algorithmic Differentiation click here.


Winner of the Prestigious Wilkinson Prize Announced


The Wilkinson Prize is in honour of the outstanding contributions of James Hardy Wilkinson to the field of numerical software and is awarded jointly by Argonne National Laboratory, the National Physical Laboratory, and NAG. It is awarded every four years to the entry that best addresses all phases of the preparation of high-quality numerical software.

The winners of the 2011 Wilkinson Prize have been announced as Andreas W├Ąchter and Carl Laird for their implementation of Ipopt, a software package for nonlinear nonconvex continuous optimization.

The prize will be awarded In Vancouver on July 21, 2011 at the International Council for Industrial and Applied Mathematics (ICIAM) Congresses. Previous winners have included Wolfgang Bangerth of Texas A&M University, Ralf Hartmann of DLR Germany and Guido Kanschat of Texas A&M University for the software package 'deal.II'; Jonathan Shewchuch for 'Triangle' and Matteo Frigo and Steven Johnson for FFTW. More details are available here.

 


NAG Toolbox for MATLAB News


The NAG Toolbox for MATLAB now supports MATLAB R2011a (7.12) release. The Toolbox gives access to the NAG Library routines from within MATLAB.

For more details about the NAG Toolbox, and the platforms supported, please refer here.


Watch the use of NAG Library in action - Time Series Modelling by the use of ARIMA


A new 5 minute webinar has been recorded. The new webinar demonstrates the use of the NAG Library for time series modelling using ARIMA (autoregressive integrated moving average). This example is based on an application developed in a .NET environment.

It can be accessed via this web page: http://www.nag.co.uk/Market/webinars.asp.


Recent Blog Posts


The Addictive Allure of Supercomputing
Performance and Results
Sticking to coding guidelines
Turning off logic
Aging Fortran Codes - A Crisis in Slow Motion
Good Business Analytics Conference


Out & About with NAG


International Supercomputer Conference 2011
19-23 June 2011. Hamburg
The International Supercomputing Conference (ISC) is a key global conference and exhibition for high performance computing, networking and storage. Now in its 26th year, ISC'11 will once again reunite over 2,000 like minded HPC researchers, technology leaders, scientists and IT-decision makers. A world-class exhibition on supercomputing, storage and networking vendors also awaits our visitors. We are once again supporting ISC 2011; we are one of the event's longest serving exhibitors +and you are most welcome to see us any time during your visit.

Quant Congress USA
12-14 July 2011. New York
Quant Congress USA, will reveal the most cutting-edge research and innovations across the quantitative analysis, trading, risk management and investment markets. Bridging the gap between theoretical perspectives and practical challenges, this annual event is widely recognized as the meeting point for top practitioners and leading academics in the quantitative finance community. It will shed new light on the models and strategies gaining importance in challenging market conditions. We will be exhibiting once again this year, and look forward to welcoming you.

HECToR / High Performance Computing Training Courses

Details of how to attend can be found on the HECToR website http://www.hector.ac.uk/cse/training/eligibility/

23-24 May 2011, NAG Oxford
An Introduction to CUDA Programming - FULL

25-26 May 2011, NAG Oxford
An Introduction to OpenCL Programming - FULL

25-26 May 2011, University of Bath
OpenMP

14-16 June 2011, University College London
Parallel Programming with MPI - FULL

For more information on any of the above events visit NAG's ‘Out & About’ webpage


New NAG product implementations


The NAG C Library, Mark 9 is now also available for the following platform:

  • Linux 64 Intel C++ 64 bit

NAGNews - Past Issues


We provide an online archive of past issues of NAGNews. For editions prior to these, please contact us.