Quantum networking offers the promise of enhancing everything quantum computing has to offer – from the possibility of faster, more accurate computation abilities to simulations that may aid commercial industries like finance, commerce, transportation, and more.
Quantum computers – which rely on quantum bits, or qubits, for information exchange – leverage principles of quantum mechanics to initiate incredibly fast communication. That occurs through the manipulation of qubits (which, through a property called superposition, can exist in multiple states until they’re measured) through lasers and electromagnetic fields.
But, what’s the catch?
Factors like noise, state preparation, and hardware and measurement complexities can hinder the performance of today’s quantum computers, particularly their scalability as more qubits are added to a single system. Between their hardware, environmental requirements, and wiring, certain high-powered quantum systems also require large amounts of space.
However, the key to meaningful (and sustainable) performance gains lies not in filling stadium-sized spaces with hardware and rigid wiring, but in making efficient quantum networking technology work – a concept that has long been a focus for IonQ and its founders. In fact, modularity in quantum computing carries potential manufacturing and commercial advantages over more monolithic machines, which can prove powerful but are not as unreliable.
What exactly is quantum networking?
Classical supercomputers are made by putting multiple cores on a single processor and distributing applications between cores and processors. Networking helps those various processors communicate. For quantum computers to scale significantly, they’ll likely have to follow a similar model – built from multiple, interconnected processors.
These devices, however, use the network for computation, not communication, effectively absorbing the computing power of separate systems versus overloading a single device. This workload distribution requires specific tools and technologies.
Quantum photonic interconnects are a key part of the networking chain, and rely on entanglement (a scientific phenomenon in which two subatomic particles remain intrinsically connected despite physical separation) to utilize qubits across different QPUs.
Overall, quantum networking has the potential to significantly increase computational power, with users tapping into distributed systems to pool computing power and run calculations that would take classical systems thousands of years.
A network of quantum computers, as envisioned by Midjourney AI
Why is quantum networking important?
In a race to strengthen individual quantum computers, be it 100-, 500- or 1,000-plus-qubit devices, scientists have encountered a series of issues, like fidelities, which degrade performance. Yet, efficacy in quantum computing depends on the ability to leverage that formidable computing power. So, the wider solution lies in networking, in which quantum workloads are more reasonably distributed, avoiding the limitations encountered on a lone device.
At a higher level, effective quantum networking could ultimately strengthen these computers’ ability to effect change in everything from risk aggregation to chemical simulations, among other scientific, commercial, and governmental use cases.
How can we get quantum networking working?
Some of the protocols required in multi-QPU information processing include entanglement generation and the use of multi-qubit gates.
IonQ’s technology, which is underpinned by ion traps, is particularly amenable to this process, in which photons emitted from trapped ions are entangled and, through laser and electromagnetic manipulation, distributed to different QPUs. That occurs through a repeatable process called “teleportation” – involving measurement and a new coupling of physically distanced particles.
Ultimately, entangled particles can be used for information sharing across a broader and more sophisticated quantum network.
What are some of the challenges?
With quantum networking still in its infancy, there is still a relatively slow rate of information exchange across QPUs (i.e., low entanglement rates), but specialized protocols and the use of multi-qubit gates can be used to reduce resulting bottlenecks. Easily integrated software packages now also help optimize performance.
Additionally, quantum architectures that do not leverage trapped ions often encounter heavy scalability challenges. In fact, those using superconducting qubits (which transmit information via microwave) face a process requiring extremely low-temperature (milli-Kelvin) environments to limit interference. The conversion of microwave photons to optical frequencies is also a very slow process that, currently, is not highly compatible with quantum computers.
Since specific state, measurement, and hardware complexities still present some scientific roadblocks, it’s also difficult to determine with any certainty the exact range of possibilities of something like a functional 10,000-qubit computer, enabled by quantum networking, would be capable of achieving.
Now, what are some of the impressive benefits of quantum networking?
Quantum computers may give us the ability to run chemical simulations, more effectively encrypt information or comb through databases faster than ever before. Its outputs have already helped auto manufacturers understand the chemical composition of batteries for electric vehicles, and will prove instrumental in running Monte Carlo simulations to better understand financial probabilities.
In time, however, our quantum needs will outgrow single machines, so continued progress will largely depend on functional quantum networks. This should provide users with increasingly insightful data in areas like commerce and the supply chain, law, insurance, and even across fraud detection/prevention, among other applications.
Quantum networking will also drive helpful cost improvements industry-wide – that is, it may become more cost-effective to purchase several lower-qubit quantum computers, for instance, and periodically network with a higher-qubit computer when necessary, versus investing in a lone, max-power device. In effect, this accessibility will also help broaden the use of quantum computers over time.
Altogether, to consider successfully, and sustainably, scaling this technology, we’ll need to rely on quantum networking via modularity, as opposed to reinventing the technology stack each time users seek stronger computing power.
Here at IonQ, we’re excited about what the future holds! Our team is currently designing a quantum network and we hope to soon demonstrate its capabilities.