The Timeline for Production Quantum Computing

The Timeline for Production Quantum Computing

Quantum computing isn’t going to replace classical computers anytime in the future, but it will become an integrated part of augmenting existing compute resources, according to experts on this week’s episode of Fiber for Breakfast.

“You wouldn’t take an airplane to the grocery store,” said Coleman Collins, Director of Product Management, IonQ Quantum Business Solutions. “They are purpose-built tools. A quantum computer is really just a new kind of coprocessor, in the same way an ASIC, GPU, or TPU is. It has some slightly different requirements, but fundamentally 100% of the application I’m talking about will still require a quantum computer and a classic computer working together in harmony.”

IonQ expects to ship its Forte Enterprise 35 qubit quantum computer in 2024, a system designed to slide into data centers and enable companies to start developing better tools and applications to take advantage of quantum computing’s unique characteristics to solve business problems in a production environment.

“We’re racing as quickly towards this promised land called commercial advantage, or production quantum computing,” said Collins. “We’re doing that by building the best hardware in the market and a stack of offerings on top of that let us engage with our ecosystem partners who are doing applications development, consulting services, and building software platforms and middleware.”

Building this ecosystem of hardware and software tools is necessary to integrate quantum computing into data center operations, where the technology is expected to provide drastic improvements in optimization, machine learning, chemistry, and other types of modeling of the physical world, manufacturing, and financial risk.

“Quantum computers will be an amazing addition to the existing infrastructure, network, and compute that we have,” said Collins. “They will not be a total replacement… it will unlock the capability to do all sorts of problem solving you can’t do right now. We’re still working on the programming models in software, but to us, it looks like a GPU metaphor. We’ll be running quantum kernels and running threads inside of that, but it’s still working in the context of a larger application that will look much like it does today.”

Depending on the application and its needs, the payoff timeline for incorporating quantum resources today could be anywhere from two to five years and will require another hardware generation or two to reach the production stage. For more on the path to production quantum computing, listen to the latest Fiber for Breakfast podcast here.