IBM Expands Access To Quantum Computing
While several companies have research projects in quantum computing, one company has been able to deliver consistent and exceptional products over the years – IBM. Earlier this year IBM updated its offerings in quantum computing and added new managed quantum systems for select international clients. And the continual progress that IBM is making shows that quantum computing is here to stay and has an interesting future for solving complex computing problems. We also expect an update to IBM’s quantum computing roadmap at IBM Think this week.
Quantum computing isn’t perfect for every problem. It’s best used in complex problems such as simulating chemical systems or some machine learning applications. Currently, these types of problems are typically solved using supercomputers. But these problems become exponentially more difficult to compute as the number of variables increases. A quantum computer’s ability to use quantum mechanics to solve these problems in a completely new way can change these exponentially large problems to problems that scale linearly in the number of variables. The problems include areas where researchers are attempting to simulate quantum chemistry (also called quantum mechanics), material sciences, and high-energy physics. For business operations, quantum computing can help with portfolio optimizations, risk analysis, and Monte Carlo simulations. There’s also overlap with AI in areas such as model training, pattern recognition, and fraud detection. The results of quantum calculation isn’t a pure binary “yes-no” answer, but a probabilistic distribution output. To help get the most out the quantum computers with error mitigation and faster execution times, IBM is now offering Qiskit Runtime fundamental building block primitives to sample from or to estimate quantities based on this distribution output.
In April, 2022, IBM gave an updated status on its quantum computing program. IBM has over 20 operational systems accessed by over 410,000 users that perform over 3.5B executions daily. The IBM Quantum Network of partners has over 180 members, including over twenty hubs (regional centers), over ten industry joint development partners, over forty startups, and over thirty academic researchers. IBM is also building regional centers such as Fraunhofer (Germany) and the University of Tokyo (Japan) which are in operation today. More regional sites are coming online in 2023, including the Cleveland Clinic (Ohio, USA), Yonsei University (South Korea), and Quebec (Canada).
There have been over 800 publications in over 45 applications areas form groups using the IBM Quantum Network partners. Big name IBM partners include: Fraunhofer, Samsung, Boing, Dell, Goldman Sachs, BP, ExxonMobil, Deloitte, and Oak Ridge National Labs.
Market Analysis by Boston Consulting Group
The benefit of solving these problems extends beyond just the scientific value to actual creation of market opportunities. A market analysis by Boston Consulting Group (BCG) puts the added market value creation at over $3B in 2024 and continuing to grow from there. We’ve only begun to tap the potential of quantum computing.
Last year, IBM announced its “Eagle” Quantum Computer. Eagle is a 127-qubit quantum processor and currently IBM’s highest performing quantum computer. To read more about it check out the IBM blog.
However, the scale of a quantum chip (as measured in qubits) is just one of three metrics that can be used to measure the performance of a quantum processor. Performance is also determined by the quality and speed as measured by Quantum Volume and Circuit Layer Operations per Second (CLOPS), areas that IBM is also focused on improving with each generation.
Accessing Quantum Hardware
But progress in quantum computing is more than just hardware, it’s also the software support and available libraries. As IBM said in its blog post: “The programming and execution model for the next phase of quantum computing needs to be fast, efficient, easy to use, and scalable.”
IBM started by building a large community of users that adopted its open source Qiskit environment. In 2021 the company improved execution speed and improved modularity with containerized Qiskit Runtime. With Qiskit Runtime, users can bundle quantum and classical execution together into programs that run on classical compute infrastructure that is co-located with IBM Quantum systems. IBM changed the execution model from a service that executes circuits sequentially, which required long communications trips to the user’s computer and back to the Quantum computer. With Qiskit Runtime, the users set up programs that run close the quantum computer, eliminating performance bottlenecks. IBM Quantum delivered 120x speedup of quantum chemistry algorithm workloads with Qiskit Runtime.
For the foreseeable future, quantum computers will be accessed through the cloud. It’s pretty much a Quantum-as-a-Service (QaaS) model. Up until now, IBM has had basically two quantum computing engagements: one is free for students and people experimenting and learning about quantum. The other has been a (pricey) institutional-level commitment with IBM to gain access to premium offerings, up to the very latest systems.
But now IBM has added a third option called “pay as you go” to reach a broader community of developers and researchers. This new plan targets experts looking for high-end quantum systems but do not have the corporate purchasing power to sign multiyear deals. This could include academics and smaller research companies. There’s no minimum spend in the pay-as-you-go plan but you do have to sign up for a free IBM cloud account. Offerings on the pay-as-you-go plan are currently only available on 27-qubit machines. Users can develop demanding quantum programs using Qiskit Runtime primitives and execute them on IBM quantum systems at a rate of $1.60 per runtime second.
The free open plan is still available, but it’s limited to 5-qubit quantum computers and below, but it’s a great place to get started and for entry level quantum developers.
IBM has consistently delivered what it has promised on its aggressive quantum computing roadmap and the company has no plans to slow down. The IBM Quantum Development roadmap update at IBM Think, will be the subject of a follow-on article.
Tirias Research tracks and consults for companies throughout the electronics ecosystem from semiconductors to systems and sensors to the cloud. Members of the Tirias Research team have consulted for IBM, Nvidia, Qualcomm, and other companies throughout the AI and Quantum ecosystems.