The Other P is Destroying the Ability to Achieve Performance Portability

An important consideration when moving to the next generation of multi/many-core systems is striving to create a refactored application that can run well on available systems. There are numerous similarities between the multi/many-core systems of today. Today’s systems have very powerful nodes, and the application must exploit a significant amount of parallelism on the node, which is a mixture of MIMD (multiple instruction, multiple data) and SIMD (single instruction, multiple data). The principal difference is how the application utilizes the MIMD parallelism on the node. Multicore Xeons and many-core Intel Phi systems can handle a significant amount of MPI on the node, whereas GPU systems cannot. There is also a difference in the ... [ Read More ]

Cray Swiss Clients Receive Innovation Award

As we highlighted at the recent SC16 conference, Cray’s clients do amazing things that affect ordinary people every day. So it was nice to see MeteoSwiss (Switzerland’s Federal Office of Meteorology and Climatology) and the Swiss National Supercomputing Centre (CSCS), recognized for doing just that. The Swiss ICT Award 2016 honors MeteoSwiss and CSCS for their work on the COSMO Novel Expert Tool (COSMO-NExT). The award website explains: “The demands on the national weather service are continuously increasing. Weather simulations are essential for flight safety, storm warnings and in the event of nuclear or chemical disasters. In order to achieve accurate forecasts and create more realistic simulations, new and higher resolution ... [ Read More ]

The Cray® XC50™ and NVIDIA® Tesla® P100 GPU – the Next Giant Leap in Compute Performance

Europe’s fastest supercomputer shoulders the challenges of the region’s toughest compute problems, from weather modeling to cosmology, and across a very broad range of other scientific domains. As the computational demands for this popular resource increase due to finer resolution on climatology models, more complex scientific computations, bigger datasets, and larger simulations, and as workloads diversify, the system needs to increase in capability and evolve while maintaining a fully utilized production environment. Based on challenging requirements like these, Cray developed the XCTM series performance computers to easily support differing processor technologies in the same architecture. By leveraging the advantages of continuously ... [ Read More ]

A Super Interconnect for a Supercomputer

  It’s hard to truly understand the capabilities of modern-day supercomputers, their constituent technologies and deeply evolved software stacks. Once in a while, a technologist (or a marketing person) will refer to the "Cray inside my cellphone" or compare IBM’s 1980s DASD storage (pronounced dasdee by old-timers) with today’s solid-state storage devices (likely found inside a cellphone as well), but those comparisons are typically made to marvel at today’s cellphones rather than to marvel at today’s supercomputers. (In fact, today’s supercomputers now have aggregate processing and storage capabilities that would easily translate to millions of those cellphone chips and SD cards.) Impressive as that “millions of cellphone ... [ Read More ]

Chapel Completes First “Summer of Code”

The Chapel team recently concluded our first Google Summer of Code (GSoC), which we introduced in our spring blog post. GSoC is a program in which Google funds students around the world to spend their summers working on open-source projects. The process began with Chapel applying and being accepted as a mentor organization back in February. After mentor organizations were announced, students began working with the accepted organizations to lay the groundwork for their project proposals, which were submitted in March. Google provided Chapel with two student slots, a typical number for a first-year mentor organization. This made choosing students among the pool of 46 applicants challenging, as there were well over a dozen highly impressive ... [ Read More ]