What is quantum computing?
This article explains the principles of quantum computing, how it compares to classical computing, and how it uses the principles of quantum mechanics.
History of quantum computing#
Quantum systems, such as atoms and molecules, can be difficult or impossible to simulate on a classical computer. In the 1980s, Richard Feynman and Yuri Manin suggested that hardware based on quantum phenomena might be more efficient for the simulation of quantum systems than conventional computers.
Quantum states grow exponentially#
Consider a system of particles with 40 possible locations where those particles can exist. The system could be in any of 2^40 unique states, because each location can either have or not have a particle. If these are classical particles, the system is always in only one of the 2^40 states, so a classical computer needs only 40 bits to describe it. But if these are quantum particles, the system exists in a combination of all 2^40 states — a classical computer would need to store 2^40 numbers, requiring over 130 GB of memory. A quantum computer needs only 40 quantum bits to describe this same quantum system.
Add one more location and the number of unique configurations doubles again. At a few hundred particles, the memory required to store the system on a classical computer exceeds the number of particles in the universe. There's no hope of simulating quantum dynamics for larger systems on conventional hardware.
Turning difficulty into opportunity#
This observation posed a powerful question: if quantum systems are difficult to simulate on regular computers, what happens if we build a machine that uses quantum effects for its fundamental operations?
In 1985, David Deutsch showed that a quantum computer could efficiently simulate the behavior of any physical system — the first indication that quantum computers could solve problems too difficult for classical computers. In 1994, Peter Shor discovered a quantum algorithm to find the prime factors of large integers, running exponentially faster than the best known classical algorithm — a discovery that sparked huge interest in quantum computing, since it could potentially break cryptosystems like RSA and Elliptic Curve Cryptography.
Since then, efficient quantum algorithms have been developed for searching unordered databases, solving systems of linear equations, machine learning, and simulating physical systems in chemistry, physics, and materials science.
What is a qubit?#
Just as bits are the fundamental object of information in classical computing, qubits (quantum bits) are the fundamental object of information in quantum computing.
Classical bits are binary and can be in only one of two states, 0 or 1, at any given time. Qubits, though, can be in a superposition of both the 0 and 1 states at the same time — there are infinite possible superpositions of 0 and 1, and each is a valid qubit state. For example, 8 regular bits can encode up to 256 unique values but only represent one at a time; 8 qubits can encode all 256 values at the same time, because they can be in a superposition of all 256 possible states.
For more detail, see The qubit in quantum computing.
What are the requirements to build a quantum computer?#
A quantum computer uses quantum systems and the properties of quantum mechanics to solve computational problems: the qubits, the interactions between qubits, and operations on the qubits to store and compute information. Popular qubit technologies include trapped-ion qubits, superconducting qubits, and topological qubits — some require temperatures near absolute zero to maximize coherence and reduce interference.
The five criteria for a quantum computer#
A good quantum computer should have these five features, known as the Di Vincenzo criteria:
- Scalable — it can have many qubits.
- Initializable — it can set the qubits to a specific state (usually the 0 state).
- Resilient — it can keep the qubits in a superposition state for a long time.
- Universal — it doesn't need to perform every possible operation, only a universal set from which any other operation can be decomposed.
- Reliable — it can measure the qubits accurately.
Building devices that meet these five criteria is one of the most demanding engineering challenges ever faced by humankind. Azure Quantum offers a variety of quantum computing solutions with different qubit technologies from partners like IonQ, Pasqal, Quantinuum, and Rigetti.
Understand quantum phenomena#
The two most important quantum phenomena that differentiate quantum computing from classical computing are superposition and entanglement.
Superposition#
Imagine turning all the way to your left, then all the way to your right — you can't face both directions at once. A quantum particle, however, can have a certain probability of facing left and a certain probability of facing right at the same time, due to a phenomenon known as superposition (also called coherence). Only quantum systems like ions, electrons, or superconducting circuits can exist in the superposition states that enable the power of quantum computing.
Entanglement#
Entanglement is a quantum correlation between two or more quantum systems. When two qubits are entangled, they're correlated and share the information of their states such that the quantum state of individual qubits can't be described independently — you can only know the quantum state of the global system, not the individual states. This correlation persists even when the qubits are separated over large distances: measuring the state of one qubit provides information about the state of the other, which is very useful in quantum computing.
To see entanglement in code, see Create your first Q# program.