When somebody begins a keynote talk by telling you to abandon your common sense, you know you’re going to be in for a ride.
This was Dr Joseph Reger’s first request when he took the stage at Fujitsu Forum 2017.
He was there to discuss something that could fundamentally transform what’s possible in technology innovation: quantum computing.
It’s something of a mind-bending subject, but it’s in all of humanity’s interest to get to grips with it.
In this post I’m hoping to help you do just that.
Watch Dr Reger’s full keynote talk to learn more:
What makes quantum computing different?
In theory, any problem is solvable given enough time.
But therein lies the problem.
Dr Reger used the example of breaking a 2,048-bit SSL private key – one of the most secure encryption methods on the planet.
Even the fastest computer on earth would take 21 billion years to crack it.
For those trying to get their head around that number, it’s roughly one-and-a-half times the age of the universe.
If we want to solve problems on that scale we’ll have to build slightly quicker computers.
If Moore’s law continues at its current rate (which it almost certainly won’t) we’d eventually reach that stage, but where would be the fun in waiting that long?
Quantum computing, if we can achieve it, would allow us to bypass centuries of evolution.
And that’s for one simple reason:
Rather than testing multiple variants and possibilities of a problem one-by-one, quantum computing lets you examine every possibility at the same time.
How does it work?
Ordinary computers work on 0s and 1s, and no matter how clever an algorithm is it essentially just manipulates strings of bits that are either 0s or 1s.
But ‘either or’ is not a phrase you’ll hear in quantum computing. It’s based on the idea that tiny particles like electrons or photons can simultaneously take on multiple states or be in several places at once.
This is what’s called a ‘superposition’, and creating a computer that uses particles that can exist in a superposition will free us from traditional binary restraints.
These particles would represent ‘qubits’, which unlike ordinary bits could take on the value of 0, 1 or both of them at the same time.
But there’s another element to quantum computing: entanglement.
This describes the way qubits are strung together and the correlations between them, and the maths is that for every n qubits there are 2n correlations.
To put that in context: to describe a system of 300 qubits you’d need more numbers than there are atoms in the universe.
Imagine, then, the computing power that just a handful of qubits could produce.
This combination of the superposition and entanglement leads to an incredible speed-up in computing power, which Dr Reger refers to as ‘quantum supremacy’.
What does it mean for technology companies?
The big question here is: why does all of this matter?
Put simply, it’ll enable us to do things in a day that would previously have taken longer than any of us will live.
This opens up a huge opportunity for early adopters of quantum computing to accelerate their speed of innovation and leave their competitors a long way behind.
But more importantly, it could help solve some of the world’s greatest problems.
Imagine applying quantum computing to data analytics and artificial intelligence (AI) – we could discover new patterns in ways we couldn’t have hoped to achieve before.
Think about what this would mean for anything from improving cybersecurity to detecting cancer and producing personalised medicines based on a person’s atomic makeup.
In short: it would change the world. And technology companies must be at the forefront of these new discoveries.
How can we make it happen?
There are a number of things stopping commercially useable quantum computers becoming a reality any time soon.
While a 50-qubit computer is considered the target for quantum supremacy, for instance, you would need many more qubits on top of that just to run error correction on top of the algorithms.
Then there’s the issue of measurement – how do you know you’re getting the solution you desire?
Even with the most powerful algorithms you’d need to run problems several times because everything is of a probabilistic nature.
To put it mildly, these computers are very hard to build. And as a result we could be a long way off true quantum computing yet.
But there is another option…
Something that still follows the laws of quantum mechanics but is much easier to build.
It’s called quantum annealing.
The term ‘annealing’ comes from metallurgy and refers to the technique of heating and cooling a metal to alter its physical properties, reducing its hardness and making it easier to work with.
In the quantum space, we don’t use heat to create our annealing effect. Instead, we use a strong magnetic field to create a quantum tunnelling effect and then turn that magnetic field off and let the system return to a low-lying state.
Inspired by this, we’ve created our own product called the Digital Annealer in a co-creation project with 1QBit.
The Digital Annealer is a piece of hardware that uses semiconductor technology to solve combinatorial optimisation problems.
In tests, it’s solved problems 10,000 times faster than simulated annealing on a conventional processor.
It might not be quantum, but it’s certainly quantum-inspired. And once we make it commercially available it will enable us to skip 14 chip generations – that’s 30 years of development time in one innovation!
I’m really excited to see us racing towards a quantum future, even if we’re still a while off cracking true quantum computing.
Even with the annealing approach outlined above, we could see massive breakthroughs in areas like molecular or drug design, disaster management, the internet of things, AI and machine learning – all things that could transform the world and human lives for the better.
As Dr Reger put it:
“The future is brighter with quantum computing.
Latest posts by Jim Millen (see all)
- Five things to look out for at Fujitsu World Tour 2018 - July 9, 2018
- Exploring technology in a transforming Britain: an EDE in the skies - March 5, 2018
- Three tips for co-creation success from Fujitsu Forum 2017 - December 19, 2017