NAGnews | Issue 155
In this issue:
Classifying web pages (from BBC Sport website) using Non-negative Matrix Factorization
Take a look at the table below; it features a list of articles selected from the BBC Sport website.
You might notice that some of the articles are about football (at the time of writing, Liverpool had just qualified for the Champions League final), and the others are about snooker (the snooker world championship was approaching its climax as I wrote this blog). It would be trivial for anyone to open the links and categorise the articles accordingly. The question is, can we write a computer program to do this automatically? Admittedly, this is a contrived toy problem, but it touches upon the sorts of things that data analysts do every day, and is a nice way of introducing the concept of non-negative matrix factorization (more on this shortly).
Read the full blog post by Edvin Hopkins here.
Does your HPC deliver the best impact at the best cost?
NAG is the leading international center-of-excellence in business and technical aspects of HPC. NAG brings expertise, experience and impartiality to help you deliver the best HPC service, evaluate technology options, acquire HPC systems, make your application go faster, understand your competitive position, train users and managers, and ensure best impact at all costs.
Explore what help is available to guide you through the entire HPC process including:
- HPC and Cloud solution possibilities
- Setting your HPC on a strong financial footing
- Real-world HPC success stories
- Understanding how best to deliver and utilize HPC services
Webinars: Upcoming learning opportunities
Webinar: How to identify causes of poor OpenMP parallel performance using the Intel VTune amplifier | Wednesday 11 July 2018 | 15:00 GMT
This webinar is aimed at anyone who wants an introduction to using VTune to understand the causes of OpenMP underperformance. It describes a systematic way of using Intel's VTune Amplifier to identify the sources of parallel inefficiency in OpenMP code, e.g. load imbalance, serial execution, OpenMP overheads and slowdown in processor throughput.
Webinar: Large-scale application execution performance assessment | Thursday 7 June 2018 | 14:00 GMT
Parallel application performance issues change with scale and often necessitate large-scale execution measurements for their investigation. Performance measurement and analysis becomes increasingly challenging as the numbers of processes and threads get very large, as the sizes of both profiles and event traces tend to grow proportionally. It is therefore recommended to start from modest scale and then make execution configurations successively larger. Examples will be presented of using the Scalasca toolset that was specifically developed for scalable performance analysis of large-scale parallel applications.
Watch 'Performance Portability of 3D Acoustics on CPUs and GPUs using OpenMP
At a recent HPC conference NAG's Reid Atcheson, Accelerator Software Engineer, delivered 'Performance Portability of 3D Acoustics on CPUs and GPUs using OpenMP'. The talk can be view on Rice's YouTube channel here https://youtu.be/bn6d19xYv9k
Latest NAG Student Prize Winner
On Friday 20 April the Manchester SIAM-IMA Student Chapter held their annual conference. As in previous years NAG sponsors a Prize for the Best Talk. This year's Best Talk winner was Jonathan Deakin for "Optimal Coordinate Transformations for the Perfectly Matched Layer Method". Nick Dingle, NAG Technical Consultant was there to present the award.
Congratulations Jonathan from all at NAG.
Technical Report: Extending error function and related functions to complex arguments
In this short communication several extensions of the Faddeeva function are implemented using functions currently available in the NAG Library. These extensions allow the evaluation of error and related functions with complex arguments. Finally, two relevant applications employing these extensions are presented. Read more
Out & About with NAG
Exhibitions, Conferences, Trade Shows and Webinars
European HPC Summit Week Ljubljana, 28 May - 2 June 2018
Webinar: Large-scale Application Execution Performance Assessment, 29 May 2018
Key Topics include: understanding the challenges of large-scale application executions, understanding the associated performance measurement and analysis issues, best practice use of the Scalasca toolset and how it is used with some large-scale application executions - register
Fortran Modernization Workshop Leibniz-Rechenzentrum, Garching bei München, 4-5 June 2018
Fortran Modernization Workshop RWTH Aachen, 7-8 June 2018
Fortran Modernization Workshop HLRS High Performance Computing Center, Stuttgart, 12-13 June 2018
ISC High Performance Frankfurt, 24-29 June 2018
Advanced Risk & Portfolio Management (ARPM) Bootcamp New York City, 13-18 August 2018
To register with a discounted affiliate rate: go to the Packages Page ? under Courses click on 'ARPM Bootcamp' ? in the Type List select 'Affiliate' ? in the Affiliation List select 'NAG' ? click on 'Next Step.'