In this issue:
New functionality features in NAG Fortran Compiler now available for Windows and Mac
In the November 2017 issue of NAGnews we announced a host of new functionality features that are included in the latest version of the NAG Fortran Compiler. We are happy to announce that the new features are now included in the Windows and Mac versions.
To recap, the new functionality in Release 6.2 of the NAG Fortran Compiler is:
- Support and correctness checking for coarray syntax and semantics - Coarray Fortran is a parallel processing feature added to the Fortran language to aid efficient parallel programming and is scalable from single-core to multi-CPU to clusters.
- More Fortran 2008 features.
- Fortran 2003 support (complete standard including user-defined derived-type I/O), Fortran 95 and OpenMP 3.1. The NAG Fortran Compiler is available on Linux (6.2), Microsoft Windows (6.2 now available) and Mac OS X (6.2 now available) platforms. For users preferring an Integrated Development Environment (IDE) on Microsoft Windows or Apple Mac, we include the NAG Fortran Builder.
If you are an existing supported user of the NAG Fortran Compiler you can access the latest features by upgrading your software. If you have any questions about this, please don’t hesitate to contact us. Alternatively, you can try the Compiler for 30 days with a full product trial - apply here.
Second Order Sensitivities: AAD Construction and use for CPU and GPU - View the popular slides by Jacques du Toit and Chris Kenyon
Jacques du Toit, Senior Technical Consultant, NAG, and Chris Kenyon, Head of XVA Quantitative Research, Financial Markets, Lloyds Banking Group delivered: "Second Order Sensitivities: AAD construction and use for CPU and GPU" at the Fixed Income Conference in Florence in October 2017. Their presentation slides have been highly referenced online since being published - view their slides here. If you have any questions about their research or would like to know more about NAG’s Algorithmic Differentiation Solutions do get in touch.
Presentation Abstract - First order sensitivities are in common use in mathematical finance, but second order sensitivities are less common because of stability concerns, and costs of construction (compute and storage). With AAD as opposed to bumping the cost profile is radically changed. We provide estimates of second order sensitivities on CPU and GPU platforms using newly available tools. It turns out that the stability issues often result from naive Monte Carlo approximations to analytic integrals: for both AAD and finite differences, we are attempting to differentiate the un-differentiable. We discuss ways to fix this (smoothing). Given the availability of fast and stable second order sensitivities, we survey potential use cases including hedging (gamma and cross-gamma) and regulatory (FRTB, PnL explain, etc) with reference to pricing and XVA. View the slide here.
'Not only Fortran and MPI': POP's view of HPC in Europe
This article is taken from a blog by Nick Dingle, HPC Application Analyst at NAG, currently working on the EU POP Project, to which NAG is a partner.
At the recent Computing Insight UK conference in Manchester (12-13 December 2017), POP presented an overview of how we've seen people using and writing HPC software in Europe. We thought we'd summarise the talk here on our blog for those of you who couldn't make it to the event. As you'll see, there's more to the work POP has undertaken than just looking at Fortran code parallelised with MPI!
In just over two years of operation POP has undertaken over 150 investigations of codes drawn from a wide range of scientific domains. One of the first questions we ask our users is "Up to what number of cores are you satisfied with the performance of your code?". This information is useful to POP's performance analysts as we actively look to stress the codes we study in order to identify the areas that can be improved. Taken together the answers also give us some insight into how people are using HPC resources in Europe. Read the full story here.
Webinar: Benchmarking as the answer to HPC performance and architecture questions
Andrew Jones, Vice-President, Strategic HPC Consulting and Services, NAG (@hpcnotes) will deliver this impartial webinar looking at benchmarking of HPC systems. A must for anyone involved in high performance or scientific computing.
Are you trying to address any of these three questions?
- Which processor or system architecture is right for your HPC needs?
- How fast does your code run and where to optimize it?
- What do benchmarks really mean?
In this live webinar Andrew will answer these questions and discuss the topics below. There will be opportunity to have your own questions answered, either during the session or post event.
- When to use benchmarking
- What benchmarks can and can't tell you
- Rules, consistency and pitfalls
- Selecting the most appropriate benchmarks
- Extrapolating to larger scales or newer technologies
- Tips, tricks, and best practice
NAG continues to support research and students
As well as sponsoring many NAG Student Prizes at higher education level we also like to support students in other ways. We are delighted to fund a PhD student in attending the forthcoming FEniCS'18 conference at the University of Oxford, 21-28 March 2018. Do get in touch if you are interested in sponsorship opportunities or student prizes.
Best of the Blog
Using the NAG Compiler with the NAG Fortran Library (Mark 26) on Windows
The NAG Fortran Compiler is an excellent compiler for checking and running your Fortran code. We use it extensively here at NAG to ensure that our code for the library complies with the current Fortran standards.
Personally, whenever I have a user problem report that I can't resolve by inspection my first instinct is to run the user's code with the Compiler. Frequently this identifies the error immediately.
As I am a Windows user, I am able to make use of the Integrated Development Environment (IDE) for the compiler that is provided to our Windows users. We call this IDE 'NAG Fortran Builder'. One of the nice features of this IDE is its ease of use with the prevailing NAG libraries. To do this the user normally specifies a 'NAG Library Project' at creation time; thereafter the relevant settings are made to the compiler, so that either the Windows 64-bit or Windows 32-bit DLLs are used. Read more
Learning software development at NAG
My name is Philip Maybank and I came to work at NAG for 6 months on placement from PhD study in Statistics at the University of Reading. During my PhD I did a lot of coding / programming to generate results for my thesis (i.e. for my own use). In doing a placement at NAG I wanted to learn how to write code that can be used easily and effectively by other people. I also wanted to learn about how numerical algorithms are used in industry. NAG proved to be an ideal place for me to pursue these goals. Read more
Out & About with NAG
Exhibitions, Conferences, Trade Shows and Webinars
Webinar: Benchmarking as the answer to HPC performance and architecture questions, 1 March 2018 Online
SIAM Conference on Parallel Processing for Scientific Computing, 7-11 March 2018, Tokyo
Supercomputing Frontiers Europe 2018, 12-16 March 2018, Warsaw
Rice Oil & Gas HPC Conference 2018, 12-13 March 2018 Houston
SCAsia2018, 26-29 March 2018, Singapore