2. Recent software and hardware developments

As for recent software and hardware developments, we are currently in the so-called NISQ phase, the Noisy-Intermediate-Size-Quantum Computer, which is characterized by universal (can perform every calculations) quantum computers with several hundreds of operational units, so-called qubits, which are intrinsically noisy, i.e. operations with these qubits are prone to errors and must be corrected by designing specific error mitigation algorithms. Thus, among the number of qubits, the quality becomes an indicator of the performance of recent quantum computers.

Just late 2022, IBM unveiled a 433 qubit processor “Osprey” with triple as many qubits as “Eagle”, IBM’s previous processor, positioning “Osprey” as the most powerful universal quantum computer to date, carrying the capability to accelerate certain calculations millions of times faster than the fastest supercomputers. Why is that so revolutionary? The number of classical bits required to represent a state on “Osprey” massively exceeds the total number of atoms known in our universe, which is approx. 1080. It carries the potential to speed up a diverse set of processes e.g. discovering new products, improving data encryption, optimizing strategies for investments, modelling oceans or space, and finding suitable treatments for diseases, exceeding anything seen so far, and making it possible to create reality at a significantly faster pace. Recently in 2023, researchers at IBM aim to release a more than 1.000-qubit processor; within the next two to three years, the researchers target to further scale and develop processors with more than 4.000 qubits (IBM Kookaburra Quantum System) by following a new approach – the modular system “Quantum System Two”. Let’s make that more approachable: In 2019, Google’s “Sycamore Processor” with only 53 qubits - 1/75 of the number of qubits IBM’s “Quantum System Two” will have – was able to perform a calculation within 200 seconds that would have taken the fastest supercomputer 10.000 years.

The “IBM Quantum System Two” is said to be the next wave in quantum computing as it uses communication links to combine several processors in one system. More precisely, as it employs a modular architecture and quantum communication it becomes scalable and increases the computational capacity. Moreover, the system allows to seamlessly integrate quantum and classical workflows by employing hybrid cloud middleware. The goal of a modular concept is to connect millions of qubits & other components and link them with microwave connections or quantum-friendly fiber-optic to open a new door: distributed, large-scale quantum computers to achieve a new state of power as we currently have it with the power of distributed classical computing.

The key advantages of qubits are the properties of superposition and entanglement. But those properties come to a price that the qubits must be isolated from their environment and thus operates at temperatures very close to absolute zero. The qubits can be isolated just for a very short time, long enough to perform calculations but too short for performing very complicated algorithms. To counteract the effects of environmental changes, e.g., Earth’s magnetic field or radiation from cellphones, that can lead to the collapse of the superposition and entangled states, companies like Google, IonQ and IBM test several approaches. Google demonstrates error correction by outperforming underlying qubits with other qubits assembled into error-correcting sets. IonQ and IBM both take similar approaches by mitigating in a variety of ways, e. g. including an option in the API to reduce error count by trading speed, allowing users to accelerate the development of quantum applications.