From 24 to 28 June, more than 3400 members of the scientific computing community gathered in Frankfurt, Germany, for the annual ISC High Performance conference. The event showcases the latest developments in a host of fields related to high-performance computing (HPC) and features a significant industry show, with more than 150 companies and research organisations exhibiting. It also plays host to the biannual announcement of the “TOP500” list of the world’s fastest supercomputers, with Oak Ridge National Laboratory’s Summit supercomputer taking this year’s top spot following a five-year period of domination by the Chinese machines Tianhe-2 and Sunway TaihuLight.

Maria Girone, CTO of CERN openlab, gave the keynote talk for this year’s conference. CERN openlab is a unique public–private partnership between CERN and leading companies such as Intel, Oracle, Siemens and Huawei to make research carried out at CERN and other laboratories possible. Girone’s talk, titled “tackling tomorrow’s computing challenges today at CERN”, discussed the schedule of upgrades for the LHC – which will culminate in the operation of the High-Luminosity LHC (HL-LHC) in around 2026 – and how this will result in a host of new challenges (CERN Courier November 2017 p5). Using current software, hardware and analysis techniques, the required computing capacity when the HL-LHC comes online is likely to be roughly 50–100 times higher than today, with data storage expected to enter the exabyte (1018 bytes) regime. Girone ended her talk by highlighting a range of specific areas – such as machine learning and data analytics – where collaborative R&D efforts with industry are either already taking place or hold significant future potential, and social-media posts about the talk reached an audience of more than 100,000.

Members of CERN openlab’s management team, which is led by Alberto Di Meglio, held meetings with a range of existing and potential partner companies, including D-Wave, Google and Cray. The event also saw Sofia Vallecorsa of the CERN IT department awarded the prize for best research poster in the category “programming models and systems software”. Her poster presented work carried out through a CERN openlab project with Intel to explore the feasibility of using deep-learning algorithms for the simulation of particle transport in the
LHC experiments.