Microsoft speeding up its quantum computing efforts

 For most users, computer speeds are more than adequate to meet everyday computing needs. However, for a relatively small, but critical, selection of power users, the speed of the computer can be compared to that of a snail – terribly slow. The prevailing computing paradigm has been based on binary operations – zeroes and ones […]

 For most users, computer speeds are more than adequate to meet everyday computing needs. However, for a relatively small, but critical, selection of power users, the speed of the computer can be compared to that of a snail – terribly slow. The prevailing computing paradigm has been based on binary operations – zeroes and ones – from time immemorial. But as the demand for faster computers has grown astronomically in recent years – in scientific and engineering computing, computer graphics, etc. – the limitations of the binary approach has been laid bare; particularly in light of the realization that the power of computer chips might have saturated, relative to the so-called Moore’s law. Needless to say that big data analytics will be one of the beneficiaries of faster speeds.

Thus, technology companies like IBM, Google, Hewlett-Packard (HP), and Microsoft are scrambling to come up with means of achieving out-of-this-world computing speeds via quantum computing (QC). A story by Agam Shah in the PCWORD.COM online magazine on 21 November 2016 tells about how Microsoft is doubling down on its efforts to make a quantum computer (hardware and software), as computing looks to a future beyond today’s PCs and servers. According to Shah, “Microsoft has researched quantum computing for more than a decade. Now the company’s goal is to put the theory to work and create actual hardware and software.” The motivation of course is faster speed; the aim is to create universal quantum computers that can run all existing programs and conduct a wide range of calculations, much like today’s computers. Early quantum computers can be used to run only a limited number of applications. 

The idea is that we have not made much progress in improving the basic manner in which computers work – the framework of binary operations, that is. Quantum computing (QC) might just foot the bill and bring to us the much needed novelty to solve our humongous computing problems. The traditional computer does everything in terms of ones and zeroes, so that instructions and the data they operate on, must eventually be converted to strings of 0’s and 1’s. 

A quantum computer uses the laws of quantum mechanics in physics to process information. A quantum computer uses quantum bits, abbreviated “qubits,” which is a system that encodes the one and the two into two distinguishable quantum states. However, quantum particles behave “randomly” (stochastically or probabilistically, in grown-up’s terminologies). Therefore, we can exploit the fact that a quantum system can be in multiple states at the same time. In other words, something can be “here” and “there,” or “up” and “down” at the same time, in the words of the Institute of Quantum Computing (IQC) at the University of Waterloo in Canada. The ability of a quantum system to be in multiple states at the same time is referred to as superposition.

The existence of an extremely strong correlation between quantum particles is also exploited in QC. That is, two quantum particles remember each other’s locations in space and time, no matter the distance of separation – even when placed at opposite sides of the universe! This connection which, by the way, boggles the mind, is referred to as entanglement. Thus, the superposition and entanglement phenomena enable a quantum computer to process an inordinately huge number of calculations simultaneously, compared to just one at a time for traditional computers. However, coming up with the algorithms that exploit these features is a challenge. 

The potential for using quantum computers in cryptography is particularly noteworthy. While any computer can multiply two fairly large numbers, breaking a large number (say with 500 digits) into its factors is presently only in the (potential) purview of quantum computers. Thus, encryption technologies exploit the difficulty of using traditional computers to factor large numbers. The difficulty of factoring big numbers is the basis for much of our present day cryptography. It’s based on math problems that are too tough to solve. 

A precise answer to what should constitute the qubits in QC seems to be the biggest issue confronting the technology. Various groups scattered all over the world are working hard to solve this problem. Nitrogen atoms embedded in diamonds are being studied, as are photon particles – which make up light, and the use of a semiconductor. The semiconductor route will be attractive if the semiconductor can be laid on semiconductor chips already in use in traditional computers. This will shorten the time-to-market relative to that for the contending approaches. 

It does not seem as if Microsoft has launched any quantum hardware yet and, like its competitors in this endeavor, Microsoft will have to make quantum circuits on which it can test applications and tackle issues like error correction, fault tolerance, and gating. Shah points out that “qubits, which allow quantum computers to calculate in parallel, can be fragile, and interference from matter or electromagnetic radiation can wreck a calculation.” Researchers at Microsoft are believed to be working on technologies that will reduce errors.