Organizations can now try the quantum computing acceleration or emulator through cloud-based and/or on-premise systems. IonQ co-Founder and CTO Jungsang Kim shared his insights on how customers should choose the service for their own use cases.

IonQ is a quantum computing vendor developing a general-purpose trapped ion quantum computer and software. Founded in 2015, the company started building its first quantum computer in early 2017 and constructed four systems, two of those now commercially available since 2020, according to Kim.

Dell Technologies recently released a hybrid classical-quantum computing package that taps the vendor’s quantum computers to enable both on-premise and cloud-based quantum acceleration. The platform uses Dell EMC PowerEdge R750xa servers paired with IonQ’s simulation engine, quantum processing unit (QPU), Aria software, and its cloud-based quantum computing capabilities.

Other than IonQ Aria software, the vendor runs its Harmony trapped ion architecture on public clouds, and offers a 32-qubit Forte single-core quantum processor, according to the vendor. IonQ supports almost all the major public cloud providers, libraries, and tools including Amazon Web Services (AWS)’ Amazon Braket, Microsoft Azure Quantum, and Google Cloud quantum services.

During a recent interview, Kim dived deeper into cloud-based and on-premise quantum computing with SDxCentral:

SDxCentral: Why does IonQ work with almost all the major cloud providers to direct access to its latest quantum hardware?

Kim: I feel like there are many different types of users in quantum computing. Actually, we work directly with a lot of these fortune 500 companies where they actually have a problem that they are trying to tackle, which is hard. And they're looking for quantum computing services to approach it. So we actually worked very closely with them not just to provide the hardware but also to provide solutions [that, for example,] invent the algorithms for the solution and validate those in our system.

Then, there are hundreds of startup companies who are trying to create solutions for various clients in quantum, and we believe that we should enable many of those companies and researchers and enterprises to find their own solutions in quantum.

And it's so much easier to work with cloud providers and use their infrastructure and access network to provide these capabilities to those very large numbers of clients.

But sometimes I think it's a balance between working with very specific customers versus enabling a very large number of other interested industry parties and even students are one, because we need to continue to train the next generation of users and developers. So, I feel like that's a good arrangement to partner with these cloud providers, and they have their own way of educating and enabling their quantum compliance. In case they don't have high-quality products to offer, I think it makes sense for us to do that.

And I think with the Dell partnership, we can probably target a different set of customers who really are very deep into the hybrid quantum computing world where some of their problems can really benefit from having a hybrid system. So I think each of these channels to which the customers actually serve different types of customers with different kinds of purpose.

SDxCentral: Can you dig into more details on the hybrid quantum computing approach?

Kim: A lot of these business quantum applications, as we figure out what they are and how to serve that, the development of that will actually be limited by the application development environment, and I think that will inherently be the hybrid. So just getting some exposure and engagement with customers to actually understand what that is to offer them a robust platform by which they can develop their quantum solutions.

I think all of that could lead to hybrid solutions we can offer together as on-prem devices of the future and dedicated systems for their problems. At this point, I think the cloud is powerful because it allows many clients to investigate that, but once they find something interesting, they can always come to us. And then we can start to offer more customized solutions support for their application.

SDxCentral: One of Dell's blogs mentioned “leveraging the capability through an on-premises infrastructure solution may provide incremental cost efficiency and data privacy.” So, how should customers choose between cloud-based and on-prem quantum services?

Kim: It really depends on the application and the customer. The cloud is basically a shared resource, so in some cases, the quantum computer can turn to work, but if there are a lot of requests coming in at the same time, you will actually have to prioritize and do things over, so there may be times when you don't get a timely response.

And we have to maximize the utilization of our computers with the price you pay. Sometimes there are delays in getting answered. And in some hybrid applications, the delay can actually slow down the efficiency of the customer.

Now, there are other cases where in secure applications, like government agencies, sometimes countries want data sovereignty, which means they don't want to send their data out of their soil. And then there are applications where you need a real-time response. Let's say somebody in the financial industry comes up with some algorithm that uses quantum computers to estimate what they need to do right away. And it's the cloud job that could be slowed down. So there are all kinds of reasons — security, proprietary, data sovereignty, these are many other reasons why certain customers may at some point prefer an on-prem solution so that they own and control it.

Photo: IonQ co-Founder and CTO Jungsang Kim     Source: IonQ