Viewpoint: UK government should pump money into computers for science

05 Nov 2019 | Viewpoint

Former science minister David Willetts says public research institutes urgently need more processing power

The UK needs to make urgent investments in computer hardware to support its hard-pressed public research institutes, says David Willetts, former science minister.  

With public research labs such as the Meteorological Office and the British Geological Survey desperately short of processing power, the country needs at least one new supercomputer and new cloud computing facilities, Willetts said.

He was speaking at Imperial College London on Monday, ahead of the publication today by the UK research funding agency UKRI, of a roadmap for science infrastructure investment.

Willetts, a member of UKRI’s board, said, “We need some serious investment in exascale computers; we need investment in data custodianship and conservation,” said Willetts, adding, “It’s no good having this investment if we’re not producing the right skills to use this new hardware. That’s why I’m particularly interested in new masters’ courses too.”

With research growing increasingly expensive and requiring complex calculations done in giant data centres, the “grim truth” is that the UK is spending less overall on research than other countries.

“You get this British perspective that competing with other countries on things like supercomputers is a vulgar pursuit. But the fact of the matter is, we’re clearly behind France and Germany. I saw one table that suggests we are even behind Poland,” Willetts said.

“We’ve basically put all our eggs in one basket – the universities,” he said. “There’s a lack of patient, long term funding elsewhere. Let’s face it, our business R&D spend is low.”

It may use more diplomatic language, but the commentary accompanying the UKRI roadmap concurs, saying the UK’s “computing requirements are expanding, but the capacity available is already severely constrained. The UK is trailing its European neighbours in terms of computing resources.”

US tech dominance

The UK currently spends 1.7 per cent of its GDP on R&D, ranking 11th in the EU behind the likes of Sweden, Austria and Germany.

The latest government commitment is to raise the UK's percentage spend of GDP to 2.4 per cent, which would require a further £20 billion from government and £40 billion from industry.

Willetts’ warnings came as France and Germany vowed to work together to build an independent cloud infrastructure to break the dominance of US tech giants.

Concern is mounting over the possible migration of sensitive corporate data to America, home of the world’s cloud-computing giants. This US market dominance puts European companies, universities and research labs at the mercy of future price increases.

Willetts argues the UK is also under-investing in “application labs” to convert research into commercial products and services.

“The focus on universities has put Britain high up the rankings for universities but we have no matching strength in application and innovation,” he said.

The former science minister, who served in the Conservative/Liberal Democrat coalition government from 2010-2014, has a number of research-linked commitments. He is chair of the Foundation for Science and Technology, chair of the genomics research centre, the Sanger Institute, chancellor of the University of Leicester, and a visiting professor at King’s College, London.

In addition to strengthening computer infrastructure, Willetts called for the government to boost funding for the country’s network of Catapult centres, set up to specialise in helping to translate academic research to products.

Catapults are driving developments in gene and cell technology, renewable energy and advanced manufacturing.

The centres are funded through the UK innovation agency, Innovate UK, part of UKRI.

Innovate UK has suffered particularly heavily from cuts in the last few years, Willetts said.

Its smart grants competition, which backs commercially viable ideas, received 2,000 applications recently, with just under half of this amount of high quality, Willetts said. “However, they can fund only about 175 of these, meaning you have almost 700 innovative companies without funding because of cash constraints.”

At the same time, Willetts says university spin-out companies are leaving UK campuses too soon. “I personally think the pre-occupation with spinouts is grossly overdone. There is a lot of research projects dressed up as companies. Too many fledglings are pushed out of the nest far too soon.”

Never miss an update from Science|Business:   Newsletter sign-up