High-performance computing (HPC) refers to systems that, through a combination of processing capability and storage capacity, can rapidly solve complex computational problems across a diverse range of scientific, engineering, and industrial fields. It relies on parallelization techniques and algorithms to synchronize these disparate units in order to perform faster than any single processor could alone. Used in industries from medicine and research to military and higher education, this method of computing allows for users to complete complex data-intensive tasks. This field has undergone many changes over the past decade, and will continue to grow in popularity in the coming years.
HPC represents a strategic, game-changing technology with tremendous implications for economic competitiveness, scientific leadership, and national security. Since HPC stands at the forefront of scientific discovery and commercial innovation, competition is growing widely, between nations and their enterprises alike.
On June 20th – 2016, China announced the availability of its next generation supercomputer Sunway TaihuLight, which clocks in at 93 petaflops for performance, making it the fastest supercomputer in the world. The new supercomputer takes this spot from the country’s other supercomputer Tianhe-2 (34 petaflops) introduced in 2013, leaving behind the US system at a distant third (17 petaflops). China for the first time placed more machines than the US in the Top500 supercomputers list, by 167 to 165.
After China replaced the US from the top slot in 2010 for the fastest supercomputer, several countries have been engaged in building a robust HPC or supercomputing infrastructure to solve next generation computing challenges. Germany has been able to take greater strides in this space; and has been continuously upgrading its line-up to achieve better performances. The South Korean government has decided to independently develop a supercomputer that is needed to bring about an “intelligent information society” based on artificial intelligence and the internet of things.
The European Commission has also recognized the need for an EU-level policy in HPC to optimize national and European investments, addressing the entire HPC ecosystem and formally adopted its HPC Strategy on February 15, 2012. The implementation of the European HPC strategy in Horizon 2020 is supported through three pillars working in synergy to develop the next generation of HPC technologies, applications and systems towards exascale, providing access to the best supercomputing facilities and services for both industry including SMEs and academia, and achieving excellence in HPC application delivery and use. These developments are bringing in HPC to be applied across varied industries and applications.
With several countries building a robust HPC infrastructure to solve first generation computing challenges, India has also just begun the upgradation from Teraflops to Petaflops performance. However, other countries are also expected to move towards building Exaflop HPC systems in the next few years. Blueocean Market Intelligence, predicts that Exaflop performance could be achieved by 2018 (figure a), owing to the factors such as economic competitiveness, convergence of big data, new product and technology innovations for embedded processors.
HPC is helping humanity to solve some of our most challenging problems. HPC adoption and production will remain vital to countries’ industrial competitiveness, national security, and their ability to lead in science frontier. The global race for HPC leadership is intensifying as China, Japan, and the European Union vie to develop exascale supercomputers. Countries must also enact proactive policies to ensure that existing and future HPC systems reach “the missing middle” so that firms of all sizes can reap the benefits of using HPC systems. HPC has and will remain critical to their economic and industrial competitiveness.