Quantum computers for the most part remain gigantic endeavors in terms of the investment, research, and patience required to build them, but also quite literally gigantic–often room-sized in large part due to massive amounts of wiring and the big dilution refrigerators needed to keep them cool.
Because of all that, it is hard to imagine seeing them in corporate enterprise computing environments, whether on-premises or in space-confined data centers, and most quantum computer users now and in the near future are more likely to access them via a cloud service for an hourly or per-task fee.
However, College Park, Maryland-based quantum computing start-up IonQ believes that enterprise and data center deployment of quantum computers will move closer to reality over the next two years. To prove its point, the company this week announced its two latest quantum computer models: the fourth-generation #AQ35 IonQ Forte Enterprise (#AQ35 stands for 35 “algorithmic qubits” a quantification of computing power that IonQ claims best describes the usable qubits in a quantum system) and the fifth-generation #AQ64 IonQ Tempo, both of which are enclosed rackmount systems designed specifically for deployment in enterprise data centers.
IonQ CEO Peter Chapman speaking on the unveiling of the new machines at the Quantum World Congress event in the Washington, D.C., area, described the Forte Enterprise, which is housed in eight 48U rackmount cabinets (according to a photo provided by IonQ) as “40% smaller than the previous [third] generation” IonQ Forte system, but with more computing capability–64x more processing power–than that #AQ29 machine.
“It’s enterprise-grade,” Chapman said. “It fits with your standard data center so you don't need to build a new building for your quantum computer. It has all the electrical and AC specifications that are found in standard data centers today.” IonQ’s trapped-ion systems also do not require the same huge cooling towers that other systems have.
The Forte Enterprise is due to be available in 2024, and Tempo in 2025, with the latter expected. IonQ has not specified pricing for the new hardware, but the company already has sold one of each model to be deployed at Quantum Basel, a quantum and AI innovation campus in Switzerland, as part of what IonQ described as a $28 million agreement.
IonQ co-founder and CTO Jungsang Kim said IonQ machines already are proving themselves viable in machine learning applications and complex combinatorial optimization problems with companies like Hyundai and Airbus, the former of which has been using IonQ systems to simulate electrical vehicle battery chemical reaction and to test and improve object detection and recognition systems for autonomous vehicles.
“We believe in the enterprise-grade quantum computing, which is where it can be something of value for enterprises, can happen in the next few years as we build powerful enough quantum computers that can actually do things that classical computers will have a challenge of doing,” Kim said.
Boh Kim and Chapman said the IonQ Tempo will be the first quantum machine with enough computing power and accuracy to do that, and that it could be the gateway to much broader deployment.
But the company still has work to do. Chapman emphasized that IonQ is continuing to work to bring down the cost of its quantum computers, and said that the smaller form factor of the new machines will help accomplish that goal.
“Smaller is also a predictor of cost,” he said, adding that as enterprise acquire a taste for quantum computing and want more powerful machines, the sector will have to start delivering networked quantum computers, and that companies like IonQ “won’t be able to charge half a billion dollars for that.”