Issue 81, 25 June 2009
- Calling the NAG Library from Java
- NAG Numerical Routines for GPUs
- Supercomputing: When supercomputing benchmarks fail to add up
- London Quant Group Investment Technology Day
- NAG Recognises Student Excellence with NAG Prize 2009
- Ask the expert! Use the g95 compiler with the NAG Library?
- Out & About with NAG
- New NAG product implementations
Calling the NAG Library from Java
Flexibility is inherent in the NAG Library. Having the ability to call library routines from many different programming languages, packages and development environments is vital to the libraries applicability. Java is one such programming language that NAG routines can be called, enabling application builders to dramatically enhance their Java application's mathematical and statistical capabilities.
The white paper “Calling NAG Library Routines from Java” gives detailed instructions on how to call routines in the NAG C and Fortran Libraries from Java. It shows examples using Java running on both UNIX and Microsoft Windows platforms. It also illustrates how to call the new option pricing routines and global optimization routines which are now available in the latest version of the NAG Fortran Library, and which will feature in the next release of the NAG C Library.
NAG Numerical Routines for GPUs
NAG has been at the forefront of numerical algorithm provision for nearly four decades and during that time has delivered algorithms in many different guises addressing numerous application areas. One area where NAG is currently collaborating with Oxford University Professor Mike Giles, is the provision of routines optimized for Graphics Processing Units (GPUs).
GPUs have traditionally been used for 3D gaming acceleration on personal computers but are rapidly evolving for use in general purpose scientific computations and are often referred to as GPGPUs ' NVIDIA use the term ‘personal supercomputing’.
Part of the collaborative research undertaken by Professor Giles and NAG has been the development of a set of random number generators and associated numerical routines which, when used on a GPU architecture, result in spectacular performance gains compared to traditional CPU systems.
Ian Reid, Chief Commercial Officer at NAG recently presented at a NVIDIA event, where he spoke about NAG developments in the GPU area to the Finance community, which include early successful usage at BNP Paribas. Ian's presentation is available here
To learn more about NAG's work in GPUs and other research projects visit http://www.nag.co.uk/market/nagresearchoverview.asp.
Supercomputing: When supercomputing benchmarks fail to add up
Benchmarking is either an invaluable procurement tool or just a pointless attempt to quantify the immeasurable, says Andrew Jones, Vice-President HPC Business at NAG.
Benchmarking is valuable for many reasons. It can measure the performance of a system architecture or new algorithms, or evaluate progress in application development and identify hotspots to optimise. But the highest-profile use of benchmarking is to help with high-performance computing (HPC) buying decisions.
Benchmarking is so important in procurement that it would be foolish to try doing without it. Yet it is also something that can be hard to get right, because of complexity, the plethora of options and user applications, and the difficulty of keeping the benchmarking in proportion to the scale of investment.
In supercomputing, benchmarking is an area of specialism in its own right, both on the vendor and user sides. It is a discipline that can help make the most of your investment, but it can also make your head whirl in the process.
Read the full article here.
London Quant Group Investment Technology Day
NAG will be presenting ‘NAG's Optimization Routines and Applicability to Finance’ at this prestigious event on 1st July. The Technology Day is part of a series of seminars, organized by the London Quant Group, to provide a forum for the presentation and discussion of practical and quantitative investment ideas.
The event is free to attend, however previous events have been over-subscribed, so if you'd like to attend we're encouraging early registration http://www.lqg.org.uk/. Alternatively, please email firstname.lastname@example.org as NAG has provision to register up to five NAG clients, which we will do so on a first come, first served basis.
For more event information or to register direct with LQG, please visit http://www.lqg.org.uk/
On 4th May Rob Meyer presented two University of Manchester students with the NAG Prize 2009. Now in its third year the NAG Prize is given annually to recognize the achievements of students on mathematical and computing related MSc courses at the University of Manchester.
This years' winners are Rebekah Carter for her MSc in Mathematics & Computational Science and Eleni Koufopavlou for her MSc in Mathematical Finance. Both students were chosen for their outstanding results and achievements on their chosen courses.
Ask the Expert! Use the g95 compiler with the NAG Library?
Question: “Can I use the g95 compiler with the NAG Library?”
Answer: Good program developers know that it's always wise to test their code using more than one compiler, for the simple reason that different compilers can find different problems. Code that one compiler accepts without a murmur may cause another compiler to complain about a possible error.
At NAG we try to make our library products work with as many compiler and operating system combinations as possible. It takes more work than you might imagine to make this happen, and sometimes we are not able to provide customised libraries for all the compilers that our users would like.
A case in point is the g95 Fortran compiler. g95 is a compiler which runs on various operating systems, and in particular is often used under Linux.
We do not have a version of the NAG Library specifically for use with g95, but it turns out that it is possible, without too much effort, to link to the gfortran version of the NAG Library. If you already have a gfortran version of the NAG Library, and would like to try using it with g95, you can find detailed instructions on how to do it here:
Notice that you do need to have the gfortran run-time library (libgfortran) installed on your machine even though you are compiling with g95 - that's because the NAG Library has dependencies on the run-time library that need to be satisfied at link time. Answered by Mick Pont, Senior Technical Consultant at NAG.
Out & About with NAG
- NAG Talks at Queens University Belfast
30 June 2009, University of Belfast
Craig Lucas, HPC Software Developer at NAG will present, ‘The HECToR Service’ where he will discuss the HECToR and what it offers the research community in the UK. As well as describing the hardware, he will explain the application process and the support and training available to users. This will include the Distributed Computational Science and Engineering (dCSE) support mechanism that awards funding for software development for codes to be run on HECToR. His second session entitled ‘The NAG Toolbox for MATLAB’ will include an overview of the NAG Library and show how to use the NAG Toolbox for MATLAB, including demonstrations and finding help, as well as giving some functionality and performance comparisons.
- World Congress on Engineering 2009
1-3 July 2009, Imperial College London
NAG experts will be present at the congress, organized by the International Association of Engineers (IAENG), a non-profit international association for engineers and computer scientists. The conference will focus on topics in theoretical and applied engineering and computer science.
- Quant USA
14-16 July 2009, New York
NAG will be exhibiting at Quant USA. Attendees will gain invaluable insight into quantitative strategies adopted by leading financial institutions. NAG experts will be available to talk about how NAG's routines can enhance finance applications.
- HECToR (High End Computing Terascale Resource) Training Courses
Presented by the NAG HECToR Team
A full list of forthcoming HECToR Training Courses can be viewed on the official HECToR website here.
New NAG product implementations
NAG is committed to offering new implementations of its broad range of numerical and statistical software components and compilers and tools.'The following implementations for NAG products have become available since the last issue of our newsletter:
The NAG Fortran Library, Mark 22 is now available for the following platform:
- x86-32 Microsoft Windows using the Intel Fortran compiler
For full details of these and all other available implementations, visit the NAG site. Comprehensive technical details of each implementation are given in the relevant Installation and User Notes at http://www.nag.co.uk/doc/inun.asp
Managing your subscription
Please feel free to forward this newsletter to colleagues or to post it to your company's Intranet sites.
To subscribe, unsubscribe or learn more about managing your subscription, please see: http://www.nag.co.uk/NAGNews/index.asp.
NAGNews - Past Issues