The Importance of On-Premises Hybrid Classical-Quantum Computing

Dell Technologies and IonQ worked together to test a hybrid classical-quantum platform that better enables hybrid classical-quantum solutions.

Quantum computation has the potential to accelerate simulation, optimization and machine learning algorithm use cases. Increasingly, customers tell us they want to consume data and execute classical workloads in private, on-premises environments due to compliance, policy and privacy issues. A cohesive way of exploring quantum computation is through an end-to-end hybrid classical-quantum solution utilizing virtual Quantum Processing Units (vQPUs) for simulation. Hybrid quantum-classical algorithms combine quantum computers with classical computers. They’re expected to power the first useful class of applications for quantum computing, whether in machine learning, optimization, quantum chemistry or other use cases. This, in turn, enables cost effective methods of discovering use cases, upskilling current team members to meet future needs and developing more complete solutions for customers.

Dell Technologies and IonQ recently worked together to test a hybrid classical-quantum platform that leverages our Dell EMC PowerEdge R740xd server paired with IonQ’s simulation engine and quantum processing unit (QPU) to better enable the journey to hybrid classical-quantum solutions. With the platform, classical and quantum simulation workloads can execute on-premises, while quantum workloads, such as modeling larger, more complex molecules for pharmacological development, can be executed remotely on IonQ QPUs. Further, wait time for each quantum circuit execution is reduced significantly by IonQ’s reservation API.

Coupling Dell Technologies’ strength in classical infrastructure with IonQ’s coherence time, gate fidelity and scale enables the QPU to solve more complex problems, allows for better error correction to reduce the time spent using the QPU to solve problems and allows IonQ to run their QPU at room temperature, enabling placement in established data centers without exotic cooling. These specific IonQ benefits, in turn, enable customers to evaluate what approach works best for them to start on their quantum journey.

The key takeaways from the testing include the following:

    • The Dell hybrid classical-quantum platform integrates seamlessly with IonQ quantum simulation and quantum processing units. Quantum workloads can be developed with vQPUs running on Dell infrastructure and later, seamlessly migrated to be executed on IonQ’s remote QPU with minimal effort.
    • With IonQ’s reservation API, each quantum circuit, executing from the Dell hybrid classical-quantum platform, does not need to wait for in job-queues and can be injected into IonQ’s remote QPU directly. This capability greatly improves performance to leverage real QPU hardware.
    • Leveraging the capability through an on-premises infrastructure solution may provide incremental cost efficiency and data privacy. As a result, organizations can train quantum programmers more efficiently and drive to a business-critical use case faster.

Our testing with IonQ demonstrates the power of an end-to-end hybrid classical-quantum solution. We’re finding as quantum hardware, algorithms and hardware continue to evolve, the need for classical infrastructure will accelerate and scale accordingly. This is just the beginning of classical infrastructure with QPUs, creating endless possibilities.

About the Author: Ken Durazzo

Ken Durazzo is the Vice President of the Dell Research Office, where he leads a team of innovative researchers who explore cutting-edge technologies for the future of Dell and its customers. Ken has over two decades of experience as a technology leader in various domains, such as networking, security, DevOps and internet-enabling technologies. He has worked for some of the largest and most respected tech companies in the world. Some of his recent research topics include hybrid quantum computing, quantum-safe cryptography, generative AI, human-machine interfaces, heterogeneous computing acceleration and virtualization. Ken was previously a distinguished engineer at Cisco and currently holds 75 patents, with several more pending. He is the Dell representative to the World Economic Forum (WEF) Quantum Network and National Academies of Science, Medicine and Engineering GUIRR program. Ken is also a frequent speaker at industry events, where he shares his insights and expertise on a wide range of topics.