Computers have been growing more and more powerful for some time now and, as processors become faster and memory more plentiful, we're reaching the limits of current technology.
Quantum computers are the next step in this evolution and more and more companies are investing in this area. Intel recently announced it will be putting over £30 million towards the technology and just this week Google boasted that its quantum computer is 100 million times faster than a regular PC.
For a deeper insight into this growing area, we spoke to Mike Mayberry, Corporate Vice President and Managing Director of Intel Labs. The full interview can be found below.
As a taster, how does quantum computing differ from traditional binary-based computing?
Quantum computing uses physical principles to do calculations. Simplistically you prepare a quantum system with the equivalent of many possible answers and then apply operations to encourage the system to evolve towards the right answer. It is by nature a probability calculus and only a subset of problems are considered appropriate for QC.
Can you give us some concrete examples of where quantum computers could excel compared to current technologies?
Simulating the properties or shape of a molecule is by nature a quantum calculation. We can emulate that calculation with traditional binary computing but QC could do the calculation natively. There are other hard mathematical problems where QC could find a likely answer quicker than sorting through all of them sequentially.
Would they still be built on silicon? What would a quantum chip physically look like?
There are about a dozen technologies under investigation, all of which use conventional electronics to control and manipulate the properties but they look very different for the quantum elements. One of the choices that looks more scalable is quantum dots which then look much like a conventional silicon chip. If you saw a QC system most of what you’d see is cooling and control electronics.
How would Intel-based QC offerings be different from products currently on the market (D-Wave Systems)
The D-Wave system is a quantum annealer, not a general purpose QC. We would pursue the more general computing model. It is too early to say what our commercial offerings might look like as we are over 10 years away.
A couple of months ago, Intel announced a significant investment in a Dutch university. What prompted the company to choose that particular organisation? What do they have that Intel couldn't either acquire or replicate inhouse?
Delft has a broad expertise in the physics and control systems. Intel can provide value with fabrication and integration of complex systems. We think that by working together we can go faster than either working alone.
Will Quantum Computing be where storage and computing will be unified?
No. It is likely that you are thinking of neuromorphic computing which operates more like how our brains work.
Will QC extend Moore's law (which is based on transistor count) or will it make it redundant?
The opposite is true. Advances in Moore’s Law had led to more sophisticated electronics which allows more precise control of the physics and allows more complex operations. If/when QC is a reality, it will supplement conventional computing, not replace it.
Image source: Shutterstock/Danor Aharon