How Supercomputing Will Evolve, According to Jack Dongarra

Quantum computing is interesting. This is a really great region for research, but my feeling is that we have to go a long way. Today we have examples of quantum computers – hardware always comes before software – but those examples are very primitive. With a digital computer we think of a counting and getting an answer. The quantum computer is about to distribute a feasibility instead of our answer and you are going to create several, we will drive it to quantum computer, and this will give you several possible solutions to the problem, but it will not answer you. So it is about to be different.

With quantum computing, are we caught at the moment of the hype?

I think unfortunately it has been over -oversized – there is too much hype associated with quantum. The result is usually all the people will get excited about it, and then none of the promises made it will survive and then the tension will break.

We have seen it before: AI has passed that cycle and recovered. And now today AI is a real thing. People use it, it is productive, and it is about to serve a purpose in a very enough way for all of us. I think quantum has to go in the winter where people are discouraged by it, they will ignore it and then there will be some brilliant people who determine how to use it and how to make it so that it is more competitive with the traditional tactic issues.

There are many problems that need to be done. Quantum computers are very easy to disturb. They will have lots of “errors” – they will break because of how fragile the count is. As long as we can make things more resistant against this failure, it is not going to do much work that we hope it can do. I don’t think we will ever have a laptop that is quantum laptop. I can be wrong, but of course I don’t think this will happen in my lifetime.

Quantum computers also need quantum algorithms and today we have very few algorithms that can be effectively run on quantum computers. So quantum computing is childhood as well as infrastructure that will use quantum computers. So quantum algorithm, quantum software, all the techniques we have is very primitive.

When – if – if ever – can you expect transformation from the traditional to the quantum system?

So today we have many super computing centers around the world and they have very powerful computers. They are digital computers. Sometimes digital computer increases with something to increase performance – an accelerator. Today those accelerators are the GPU, the graphics processing unit. The GPU does something very good, and it simply does the thing, it is architect to do it. In the old days, it was important for graphics; Today we are refacting so that we can use a GPU to meet the few calculated we need.