The race to expand HPC infrastructure, including exascale machines, to advance national and regional interests ratcheted up a notch yesterday with announcement that seven European countries – France, Germany, Italy, Luxembourg, Netherlands, Portugal and Spain – signed an agreement to establish EuroHPC. It calls for “acquiring and deploying an integrated world-class high-performance computing infrastructure…available across the EU for scientific communities, industry and the public sector, no matter where the users are located.” The announcement was made by the European Commission.
“High-performance computing is moving towards its next frontier – more than 100 times faster than the fastest machines currently available in Europe. But not all EU countries have the capacity to build and maintain such infrastructure, or to develop such technologies on their own. If we stay dependent on others for this critical resource, then we risk getting technologically ‘locked’, delayed or deprived of strategic know-how. Europe needs integrated world-class capability in supercomputing to be ahead in the global race,” said Andrus Ansip, European Commission vice president for the Digital Single Market in the official release.
The EU, of course, is no stranger to HPC pursuit with programs such as Horizon2020 and PRACE2 the follow-on to PRACE (Partnership for Advanced Computing in Europe) among others. How they all fit together isn’t immediately clear. What seems clear is regional and national competitive zeal over HPC is rising. Earlier this week the U.K. announced plans to establish six new HPC centers. The U.K. is in the process of exiting the European Union (see HPCwire article: UK to Launch Six Major HPC Centers).
The official EU announcement characterized the initiative as “a European project of the size of Airbus in the 1990s and of Galileo in the 2000s.” Here’s an excerpt from the agreement, first signed last week in Rome:
“The participating Member States:
- Agree to work towards the establishment of a cooperation framework – EuroHPC – for acquiring and deploying an integrated exascale supercomputing infrastructure that will be available across the EU for scientific communities as well as public and private partners, no matter where supercomputers are located.
- Agree, in the context of EuroHPC, to work together and with the European Commission to prepare, preferably by the end of 2017, an implementation roadmap for putting in place the above-mentioned exascale supercomputing infrastructure that would address the following:
1. The technical and operational requirements and the financial resources needed for acquiring such infrastructure
2. The definition of appropriate legal and financial instruments for such acquisition.
3. The procurement processes for the acquisition of two world-class pre-exascale supercomputers preferably starting on 2019-2020, and two world-class full exascale supercomputers preferably starting on 2022-2023.
4. The development of high-quality competitive European technology, its optimization through a co-design approach and its integration in at least one of the two exascale supercomputers.
5. The development of test-beds for HPC and Big Data applications for scientific, public administration and industrial purposes.”
The agreement also calls for pan-European involvement:
- “Invite the European Commission to participate in this endeavor and work together on how it can be best supported at EU level.
- Agree that the target exascale supercomputing infrastructure will address the growing needs of the scientific community, and to look also for ways and conditions to open the availability of this infrastructure to users from industry and the public sector, while guaranteeing the best use of the infrastructure for scientific excellence and an innovative and competitive industry.
- Agree to enable the development of applications and services, for example those proposed in the IPCEI on HPC and BDA.
- Invite all Member States and Associated Countries to join EuroHPC.”
It’s noteworthy that similar concerns are being expressed in the U.S with regard to HPC leadership, particularly in light of Trump’s proposed budget.
Speaking with HPCwire, William Gropp, acting director of the National Center for Supercomputing Applications, noted, “It would be great if there were more than one NSF track one system. It would be great if there were more than a handful of track two systems. If you look at Japan, for example, they have nine large university advanced computing systems, not counting the Flagship 2020 system in their plans, and that, honestly, is more than we’ve got. So there is a concern we will not provide the level of support that will allow us to maintain broad leadership in the sciences. That’s been a continuous concern.” (see HPCwire article, Bill Gropp – Pursuing the Next Big Thing at NCSA)