Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
Italian energy giant Eni is firing up the world’s most powerful supercomputer outside the US this Christmas as it races rivals to build the technology infrastructure needed to better explore for new sources of oil and gas.
Built at a cost of more than €100mn, Eni’s new machine, HPC6, will be switched on in the small Italian town of Ferrera Erbognone, population 1,140. It contains almost 14,000 AMD graphics processing units: high-powered chips used to perform complex calculations and run artificial intelligence processes.
The supercomputer took fifth place in an annual list of the world’s fastest computers last month with a benchmark speed of 477 petaflops per second, behind three US research computers and Microsoft’s cloud-based Eagle computer.
Its job will be to crunch data to discover new oil and gas reservoirs, as well as to perform calculations to advance clean energy.
Lorenzo Fiorillo, the head of Eni’s research and digital department, said the computer was almost nine times faster than its predecessor and that Eni was one of the few oil companies that has kept building its own machines, rather than switching to buying cloud computing services.
“A lot of the other companies realised it would be more efficient to rent time on someone else’s supercomputer,” said Rob West, an analyst at Thunder Said Energy, adding that Exxon, Shell and Chevron had used supercomputers at the US National Center for Supercomputing Applications and the Los Alamos National Laboratory.
Eni’s supercomputing division has helped the Italian company burnish its reputation for oil and gas exploration. “We were able to find oil in places where we saw nothing,” said Fiorillo. He noted that as well as computing power, Eni had built significant capability in coding the algorithms for HPC6 to run. “We started producing our own code in the 1980s,” said Fiorillo.
“We have used the supercomputer in all of our latest discoveries,” he added, saying that extreme computing power helped Eni navigate the so-called pre-salt layer, a series of geological formations beneath thick layers of salt on both sides of the south Atlantic. “Our algorithms can create clear images of where the oil is and how big it is,” he said.
While oil companies have been using supercomputers to interpret seismic data and model the behaviour of oil and gas reservoirs for years, they are now also increasingly using AI to do everything from creating digital twins of their assets to generating hundreds of different options for how to drill oilfields and where to place their wells.
Fiorillo said that his research team now spent 70 per cent of its time on clean energy and that HPC6 would be used to research how to manage plasma clouds in nuclear fusion reactors for discovering new materials; to increase the efficiency of devices that capture carbon emissions; and to work out how to make better solar panels.
The company declined to comment on whether supercomputers would soon be overshadowed by giant computing systems, such as the Colossus data centre created by Elon Musk in Memphis, Tennessee. But it said its own data centre in Ferrera Erbognone, just over 25 miles south-west of Milan, was “well-positioned for upcoming expansions”.
