Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Allinea gets the bugs out of the British weather

Share this on social media:

That perennial topic of British conversation, the weather, may become a little less unpredictable, thanks to debugging software from Allinea. And in a second British application, the UK’s national supercomputing service, Archer, has announced that it has deployed the Allinea MAP profiling tool to tackle scalability and performance on its 2.5 Petaflop Cray XC30 system.

The Met Office, one of the world’s leading weather forecasting and research organisations, is rolling out the Allinea DDT debugger to help develop forecasting software for its new Cray XC40 system. The system will be among the largest in the weather community when fully operational in 2017. Met Office software has global reach as many institutions around the world use it in creating their own forecasts.

The new supercomputer will increase spatial resolution but also complexity of the science. So, according to Dr Paul Selwood, manager of HPC optimisation at the Met Office: ‘Developing for a system on this scale means it is more important than ever to give the developers and scientists access to tools that can handle the massive concurrency.’ Allinea’s debugging software will be used by software teams within the Met Office’s headquarters and at several partnering research universities in the UK.

Meanwhile, at the opposite end of the country, the Archer system, which already uses Allinea’s debugger at extreme scale, is adding Allinea MAP. ‘As core counts have increased, so has the need for tools that can handle these levels of parallelism. Allinea MAP provides us with a scalable way to visualise performance bottlenecks and give hints for optimisation,’ said Mark Parsons, executive director of the Edinburgh Parallel Computing Centre (EPCC). The EPCC hosts and manages the Archer system. With 118,080 Intel Ivy Bridge cores, the system is jointly owned by two UK Government-funded research councils and provides scientists with a computational capability that they cannot get with their own university systems.

Andy Turner, leader of the Archer computational science and engineering, pointed out that ‘Running applications at these larger scales often exposes issues that are not visible at smaller scales. We evaluated Allinea MAP and found it complemented other profiling tools available on Archer.’ Turner’s team works with scientists to ensure their applications are able to use the system efficiently.