Kx Systems was founded in 1993 to address a simple problem: the inability of traditional relational database technology to keep up with escalating volumes of data. This problem is even more acute today, where data records, especially in the financial industry, number in the trillions. To provide users to fast data, the founders of Kx have developed kdb+ and vector-processing programming language 'Q'. Kdb+ has the ability to quickly process data and for more complex analysis it has access...

NAG provides VB headers for all the routines in the library allowing them to be called easily from Microsoft Excel given a little knowledge of Visual Basic. Along with the headers, some Excel examples are distributed with the product. You may also find demo worksheets for a variety of routines on the NAG web site. Combining the ease of use of Excel and the power of NAG routines is a great way of creating simple, flexible interfaces for solving difficult problems.  Typically you...

Do you remember those episodes of Friends that consisted entirely of flashbacks - presumably the cast were too busy with other duties that week so couldn't film, or the writers too busy to think up a real story ... The NAG Blog has been running for just over 3 years now. In that time we have had 118 posts on a wide variety of numerical computing topics including HPC, .NET, customer engagement, GPUs, optimization, and much more. The most popular posts (pageviews) have often been the most...

Written by David Sayers, NAG Principal Technical Consultant How would you like to call a NAG routine and have the precision it computes in determined by the precision of the arguments? So, for example, your program might contain: r = s15adf(x,ifail) If x were single precision then a single precision would be returned to r. If x were double or quadruple precision then the corresponding double or quadruple computations would be performed. I can imagine a number of scenarios where this...

Last year I wrote a blog post about NAG’s work on parallelising the computation of the matrix square root. More recently, as part of our Matrix Functions Knowledge Transfer Partnership with the University of Manchester, we’ve been investigating parallel implementations of the Schur-Parlett algorithm [1].   Most algorithms for computing functions of matrices are tailored for a specific function, such as the matrix exponential or the matrix square root. The Schur-Parlett algorithm is...

A bit of background… NAG was established as a not-for-profit organisation and remains so, which means it has no shareholders or investors to answer to. All surpluses made are ploughed back into R&D. NAG is ‘owned’ by its members of which are people that founded NAG, current and past employees, collaborators and friends of the company. This all serves to make NAG a pretty unique place to work which might be a reason for the longevity of serving staff (NAG also happens to be an Investors in...

I've written in this blog before about the problems of Wandering Precision - where the results computed by a program are not consistent, even when running exactly the same program on the same machine with the same data several times in a row. At SC12 in Salt Lake City a couple of weeks ago I took part in a "Birds of a Feather" session organised by Intel, where problems like these, associated with bitwise reproducibility of results, were discussed. On the panel, apart from myself, were...

London 2012 Olympic Park I really didn’t expect to love the Olympics as much as I did. I’m not really into sport and have never taken much interest before, but this time I was absolutely enthralled. London, being the host city, obviously had a bearing on my enthusiasm and enjoyment, but what now the games are over? A lot of talk centred around the Olympics 2012 ethos of ‘inspiring a generation’ and this got me thinking about NAG’s legacy and how we too can inspire. NAG has had a busy year as...

I had a bit of a rant in my article published at HPCwire this week - “Chasing1000x: The future of supercomputing is unbalanced”. The gist of my rant is that the supercomputing community pays great attention to the next factor of 1000x in performance – and I firmly agree that next 1000x is highly valuable to the HPC community and the wider economy. But, we should give equal attention to 1000x in other areas, notably ease-of-use and growth of the user-base. And, critically, give equal peer...

In my last post we saw running the NAG serial library on Amazon's EC2. We noticed that the CPU utilization was not anywhere close to the capacity I had on the Cloud. Below, I have loaded the NAG SMP and Multicore Library on Amazon's EC2 in hopes of utilizing all the virtual cores. Routines Tested The EC2 instance used was a High-CPU, c1.xlarge instance (7GB of memory and 8 Virtual Cores). The SMP Library contains 204 tuned and a total of 337 enhanced routines for use on Multicore machines...