The IBM Quantum data center timeline
- 2019: 6 systems of up to 53 qubits
- 2020: 12 systems of up to 65 qubits
- 2021: 17 systems of up to 127 qubits
- 2022: 21 total systems, including multiple systems with 127-qubit “Eagle” processors
The IBM innovations we showcased for the president today in Poughkeepsie are vital to the future of the world’s business, industry, and society, and they will greatly benefit from the CHIPS and Science Act. The new law will help fuel the future of quantum computing by accelerating research, expanding the quantum supply chain, and providing more opportunities for researchers to explore business and science applications of quantum systems.
At IBM, we are committed to advancing the quantum industry as quickly as possible. At the center of IBM’s campus in Poughkeepsie, NY is the world’s first quantum data center, which we continue to expand at a rapid pace, catering to the growing community of quantum computing users in industry, academia, research and national labs.
The center, which opened in 2019, is now home to more than 20 quantum systems executing more than 2 billion circuits daily. It is the world’s largest fleet of quantum computing systems for commercial and research activity.
A key to the success of the data center is the IBM Quantum System One. This system is built to take quantum computing out of the lab and into a commercial environment. The stability, reliability, flexibility, and scalability enabled by this design has allowed us to quickly move from the 27-qubit “Falcon,” to the 127-qubit “Eagle” processor, while maintaining high availability.
Created at the frontiers of multiple emerging technologies, IBM Quantum System One is the world’s first integrated quantum computing systems designed for commercial use. IBM Quantum System One supports our 27-qubit “Falcon” processor, our 65-qubit “Hummingbird” processor and 127-qubit “Eagle” processor. Read more.
In addition to challenges associated with maintaining uptime in classical data centers, the analog nature of quantum computers adds additional complexity to monitoring qubit quality and ensuring the high fidelity of all qubits. Our teams have devised techniques for automated detection and resolution of issues that may arise over time with regard to reliability and reproducibility of programmable multi-qubit operations. The result: our systems can manage quantum computational research over the cloud with 99% uptime.
Since the launch of our roadmap in 2020, we have used our agile hardware development process to fabricate and iterate on the families of quantum processors — the “Falcons,” “Hummingbirds,” “Eagles,” and more — that are validated and integrated into quantum systems made available to the community. Similar to the evolution of semiconductor nodes, the industry continues to pioneer new chips while delivering on previous nodes.
Correspondingly, our team iterates and improves the performance of our quantum processors multiple times a year. We are already planning the expansion of the data center to introduce the next generation of systems, including the 433-qubit “Osprey,” to be announced later this year, and next year’s 133-qubit “Heron,” which features control hardware that allows for real-time classical communication between separate processors — enabling circuit knitting techniques. Co-located classical infrastructure for error suppression and error mitigation techniques — along with Qiskit Runtime software capabilities — significantly enhances the performance of these systems.
Circuit knitting techniques allow us to partition large quantum circuits into subcircuits that fit on smaller devices, incorporating classical simulation to “knit” together the results to achieve the target answer. Read more.
The IBM Quantum data center allows us to provide our clients with cloud access to an ever-increasing system capacity, with constantly improving scale, speed, quality, and capabilities.