Qubit to the Future

Quantum computing uses the principles of quantum mechanics where subatomic level interactions produce spontaneity. Quantum computing addresses a special class of highly-complex problems where the solution is in the context of probability instead of certainty.

The most common analogy for quantum computing is tossing a coin in the air.

You know the coin will land on its head or tail, but in the air it is either and its final value is not locked until it lands, i.e., until it is measured. The basic element of quantum computing is a quantum bit, qubit. A qubit is like the coin in midair. It maintains multiple states until measured.

In classical computing, we deal in 0s and 1s and a bit can be one or the other, but not both. This type of computing delivers linear growth with processing power. Quantum computing power increases exponentially with the number of qubits. Each qubit is the equivalent of two bits that hold 0, 1, or any proportion of both 0 and 1 at the same time. This is called "superposition". This means you can do more with less.

Quantum computing has another ace up its sleeve.

Add to superposition, the concept of "entanglement", which defines the interaction between qubits. This produces results faster. When one qubit is measured, the entangled ones are also locked in. In 2019, Google ran a random number algorithm on its quantum processor called Sycamore and a supercomputer called Summit. Sycamore generated strings of 2^53 in 200 seconds. Google claimed that Summit would have taken 10,000 years. At the time, IBM said their supercomputers could do the same operation in 2 days. Regardless of these competing claims, there is no doubt that superposition and entanglement produce faster results.

Qubits are generated from quantum dots or man-made atoms. These dots are semiconducting crystals at a nanoscale that trap electrons in a 3D space using electrical fields. Quantum dots are so small that their electronic properties are controlled by the principles of quantum mechanics. Quantum dots mimic hydrogen, lithium, and sodium atoms which have 1 electron in the outer shell. The electron pointing up or down represents 0 or 1. If you catch the electron spinning between these two positions, you have superposition and multiple states. The electron is the source of the qubit.

As the number of variables and data volumes continue increasing, some problems will require the horsepower of qubits.

Quantum computing is appropriate for understanding large-scale enzyme behavior in drug discovery, finding more robust ways to predict financial risk and optimize portfolio, faster encryption and decryption of data, analyzing gazillions of water molecules to predict the behavior of hurricanes, overcoming scale errors inherent in gyroscopes used for navigation and much more. Applications of quantum computing exist in many verticals, including bio-pharma, aerospace, automotive, finance, healthcare, manufacturing, and CP&G.

IBM, Google, Intel, Microsoft, HP, along with ColdQuanta, Zapata, D-Wave, IonQ and a host of other companies are speeding forward to deliver the hardware and tools for quantum computers. IBM has recently broken the 100-qubit barrier and has the largest quantum computer in its 127-qubit Eagle processor.

For all of the promise, quantum computing has practical limitations.

When an operation provides probabilities and not precision and when entanglements are easily disturbed, it becomes difficult to control errors and verify results. Newer techniques to manage errors are enhancing the ability to transfer data from qubits before they disintegrate from interference.

The heat generated by the machines is a significant problem. Classical computers need fans to keep the ecosystem cool. Supercomputers need to be kept under 30C. Quantum computers have to be at -273C (near 0K) at the moment. That is colder than outer space! Using photons would eliminate the temperature constraint and potentially bring quantum computing to room temperature. Getting photons to interact with each other to perform operations would require nonlinear optical crystals and that technology is years away, possibly a decade.

Feverish work is underway by the tech giants and niche companies to pack in more qubits into a processor, manage error rates, reduce the temperature constraints, and bring new tools and algorithms to market.

The next few years will be fascinating to watch in this space.

Sources:

• https://www.scientificamerican.com/article/hands-on-with-googles-quantum-computer/#:~:text=Google's%20quantum%20computing%20chip%2C%20dubbed,one%20on%20the%20chip%20failed.&text=The%20new%20chip%20performed%20the,have%20taken%20Summit%2010%2C000%20years.
• https://www.mckinsey.com/business-functions/mckinsey-digital/our-insights/a-game-plan-for-quantum-computing
• https://www.zdnet.com/article/what-is-quantum-computing-understanding-the-how-why-and-when-of-quantum-computers/
• https://www.analyticsinsight.net/top-10-quantum-computing-companies-to-watch-out-in-2021/