NESUS – or Network for Sustainable Ultrascale Computing – formed earlier this year to study the challenges of ultrascale computing. Launched with about 30 European partner countries, the cross-community initiative has grown to include 39 European countries and six countries from other continents. The goal: to produce a catalogue of open source applications for large-scale complex systems two to three orders of magnitude larger than today’s systems.
Funded by the European Cooperation in Science and Technology (COST), NESUS Action seeks to establish an open European research network that pursues sustainable strategies for ultrascale computing. The first step will be to identify synergies at the cross-section of HPC, large scale distributed systems, and big data with an emphasis on programmability, scalability, resilience, energy efficiency, and data management.
NESUS is one of the largest European research networks of this type. Coordinated by Universidad Carlos III de Madrid (UC3M), NESUS will provide a meeting place to facilitate scientific collaboration into various research topics, including sustainable system software and applications. NESUS also has the goal of promoting the leadership of Europe, as well as advancing science, the economy, and society.
Although there are other efforts targeting large scale computing systems research in Europe, COST Action projects like NESUS emphasize pan-European coordination, connectivity and networking as a means of reducing redundancy and inefficiency.
The program is part of a larger effort to develop large parallel supercomputers, exaflop-class systems, by the 2020 timeframe. The solution is likely to involve datacenters with hundreds of thousands of computers coordinating with distributed memory systems.
Ultrascale computers are expected to employ both parallel and distributed computing technologies. “Ultimately, the idea is to have both architectures converge to solve problems in what we call ultrascale,” said Jesús Carretero, professor in the UC3M Department of Computer Science and project coordinator.
“Ultrascale systems combine the advantages of distributed and parallel computing systems,” observes a statement from UC3M. “The former is a type of computing in which many tasks are executed at the same time coordinately to solve one problem, based on the principle that a big problem can be divided into many smaller ones that are simultaneously solved. The latter system, in both grid and cloud computing, uses a large number of computers organized into clusters in a distributed infrastructure, and can execute millions of tasks at the same time usually working on independent problems and big data.”
When it comes to the next generation of ultrascale computing systems, it’s not just the size that is being scaled, it is the complexity as well as the requirements for energy efficiency.
“We try to analyze all the challenges there are and see how they can be studied holistically and integrated, to be able to provide a more sustainable system,” explains Carretero.
“It is the largest COST Action ever, which shows the interest that exists for it,” the professor continued.
Under the NESUS banner, some two hundred scientists from more than 40 countries are exploring what the next generation of ultrascale computing systems will be like. Nearly 40 percent of the workforce are young researchers as part of COST’s commitment to sustainability is facilitating an ecosystem of scientists for the future of the European Union.
Although the NESUS project is scheduled to conclude in 2018, the program reflects a degree of targeted R&D that is crucial to fielding useful exascale-class systems by the 2020 timeline. Such machines are expected to enable major advances in genomics, weather/climate modeling, human brain research, and countless other disciplines. You can read more about the project’s six-phase approach, here.