The world of IT that we know today will go through a radical change in the next 10 years. I have discussed this with Paul Lewis, Hitachi Vantara Global CTO for Industry, and he agrees with me that major changes in technology will be required to satisfy the exponential demands for data and information. When we say exponential change, few people can visualize the amount of change that this requires. One way to illustrate this is with distance. If I take 32 linear steps, I can move from my office to the elevator, but if I take 32 exponential steps, 2*32, I can travel to the Moon and back. That is the type of exponential change that we will require in the next 10 years.

Compute is one area where we will need to make such a change. Moore’s law is dead, and with it has died the increasing growth in compute power that we have enjoyed in the transistor age. Moore’s law was based on the principle that the speed and capability of computers can be expected to double every two years, as a result of increases in the number of transistors a microchip can contain. We have reached the limits of Moore’s law where the transistors in computers have become as small as they can possibly be. GPUs and FPGAs will only provide incremental improvements.

In order to meet the processing power that the explosion of data will demand we need to have much more compute power. As a result. computer scientist are seeking possible solutions at the atomic and subatomic level in a field known as quantum physics. It turns out that atoms do not follow the traditional laws of physics. Quantum particles like atoms, can move forward or backward in time and exist in two places at once. It is these strange behaviors that can make it possible for quantum computers to far exceed the speed and capabilities of transistor based computers.

Classical computers use ones and zeroes to do their calculations, but quantum computers use quantum bits or qubits. Quantum computers also use ones and zeros, but qubits have a third state called “superposition” that allows them to represent a one or a zero at the same time. Instead of analyzing a one or a zero sequentially, superposition describes a quantum state where particles can exist in multiple states at the same time, enabling quantum computers to look at many different variables at the same time and perform multiple tasks simultaneously. Therefore, the time it takes to crunch a data set is significantly reduced. Quantum computers can solve problems that are impossible or would take a traditional computer billions of years to solve.

Quantum computers have another characteristic that is called entanglement. Entanglement occurs when two quantum particles or groups of particles become “entangled” and become connected so that actions performed on one particle affects the other particle even when they are separated by great distances. Albert Einstein described this scientifically as “Spooky action at a distance”. Imagine what this might mean for quantum communications or a quantum internet where you can affect something remotely without sending exabytes of data. “Beam me up, Scotty” on Star Trek might not be that far off.

Quantum computers are not expected to replace classical computers. Classical computers are better at some tasks than quantum computers such as email and transaction processing. Quantum computers are great at solving optimization problems. There are several algorithms already developed for quantum computers including Grover’s Algorithmfor searching an unstructured database and Shor’s Algorithm for factoring large numbers.

We can see the benefit of quantum computers for searching exabyte unstructured databases and solving deep learning problems. Unfortunately, factoring integers is also used for cryptography, which means that a sufficiently large quantum computer could break existing encryption codes that we have today in a matter of minutes if not seconds, exposing everything that we encrypt today. You can see why national agencies are in a race to develop this capability.

The commercialization of quantum computers has made a lot of progress in the past few years. Since quantum particles are very unstable, quantum computers must operate in a very specialized environment of near zero degree temperature in order to slow down the quantum particles. While you won’t see quantum computers in a rack in your data center, they will be accessible over the cloud. In fact, Google is already recognized as the current leader in Quantum Computing. Today, cloud access to quantum computers is available for test from several government agencies, research universities, and technology companies. There are also CMOS annealing computers that can simulate some characteristics of quantum computers and do not require Zero temperatures.

If you want to hear more about the state of this technology and other technologies like DNA storage, and AI, please join Paul Lewis and I for a webinar for a look into the future on January 15 at 12pm EST.

#Hu'sPlace

#Blog

0 comments

0 views