By Stuart Crainer and Des Dearlove
The world’s biggest supercomputer is a big deal. A few years ago we talked to members of the Japanese team involved in developing a computer nicknamed K. The name was a play on the Japanese word ‘kei’ for the number 10 to the power of 16. It is a big number and it was a big build with a $1 billion development budget and over 1,000 people involved. Development began in 2007 and ended in 2012 with the K being celebrated as the fastest of the fastest.
We met the managers and leaders from the Japanese company Fujitsu involved in this huge project. It was striking how down to earth they were. The Fujitsu team were not classic Silicon Valley material. They were neither hip nor cool. There were no jeans, not even chinos. No casual wear; nothing casual at all. Indeed, when we met the project manager he looked like a typical middle-aged Japanese corporate man, wearing a suit with a cardigan to stave off the Tokyo winter chill. Ordinary people involved in an extraordinary project.
We also talked to Aiichirou Inque, a gentle, affable yet intense man. After twenty-seven years with the company we could have forgiven him for an air of ennui. Instead, when we spoke he was a ball of creative energy, excited and under pressure in equal measure. ‘At my previous company I couldn’t do the things I wanted to do. I wanted to build something by myself, not just to use it, but to build it. For me that’s what it is all about.’
And now Inque was charged with developing the K Computer then being built at Japan’s Institute of Physical and Chemical Research, known as Riken, near Osaka.
The term supercomputer was coined in the 1960s to describe a computer that is at the frontier of current computing capability. Today, supercomputers are capable of quickly performing large scale and advanced calculations that are difficult or impossible to solve with conventional computers. For example, they are used in quantum physics, weather forecasting and climate research. Supercomputers are used instead of experimentation for physical simulations such as testing airplanes in wind tunnels and research into nuclear fusion. They are also used to compute the structures and properties of chemical compounds, biological macromolecules, polymers and crystals.
The Batista heart operation is one example of the new supercomputer’s life-changing potential. The Batista operation is an experimental open-heart surgery procedure that aims to reverse the effects of remodeling in cases of end-stage dilated cardiomyopathy. Invented in 1994 by Brazilian surgeon Randas Batista, the operation involves removing a portion of viable tissue from the left ventricle to reduce its size.
Although several studies showed benefits from this procedure, studies at the Cleveland Clinic concluded that it was associated with a high failure rate. One use of the K computer that Fujitsu is exploring is the development of a heart simulator which would allow surgeons to anticipate the likely effects of the Batista procedure – to determine whether it would be beneficial to a patient and assess the likelihood of complications.
But this is only possible if Fujitsu’s engineers work closely with the doctors who are highly skilled in performing the operation. ‘The point where the computer engineer talks to the doctor is key,’ says Inque. ‘That ability to work with customers and end users is part of Fujitsu’s DNA.’ Innovation and leadership often require collaboration and listening in equal measure.
The potential benefits for mankind are huge. But so are the challenges of building the world’s most powerful computer, as Inque acknowledges. ‘The reality is that because the supercomputer is so large we will and are encountering things we haven’t experienced before or don’t really expect. That’s part of the challenge – and it’s also part of the excitement. We are really pushing the boundaries of what is possible with computing.’
Such a project is also highly inspirational for Fujitsu employees.
‘I want the young engineers working on this project to be excited and to enjoy their work,’ says Aiichirou Inque. ‘But let’s be clear: the K Computer will make the future for Fujitsu, Japan and for human beings. It will give us the ability to look at the weather of the future and there are a huge number of healthcare uses. That’s what I mean about its power to change humanity. A computer is just a big box; what’s interesting is to see it as a tool to help mankind and society.’
The current market for supercomputers is estimated to be around $9 billion and Fujitsu hopes to control 10 percent of that market by tapping into growing demand in Europe, especially in Germany, France and the UK. What makes this all the more exciting is the advent of Big Data. Collecting and making sense of large amounts of data requires big computing power. Big Data and supercomputers could be a marriage made in heaven.
‘We are already working with customers so we understand their needs and those of the eventual end users,’ Inque told us. ‘With something like a supercomputer it is often very difficult for the customers to know how to make use of the product. We have to get out and talk to them to better understand what they are doing every day and how the computer can play a role. In the end, first and foremost, the computer needs to work and, second, it needs to be heavily used. That’s how its success should be judged. Does it make a difference to human life?’
In all our conversations with the supercomputer team at Fujitsu the word leadership was never used. The greater purpose of the project was constantly emphasized. The purpose provided leadership and the job of the project leaders was simply and repeatedly to communicate that purpose for it is the creation and communication of purpose which lies at the heart of leadership.
The inside story of Fujitsu is told by Kyoko Katase and Atsushi Tajima in Fortune Favors the Brave (Nikkei, 2012).