IBM’s roadmap for building an open quantum software ecosystem
Quantum computing is on the verge of sparking a paradigm shift. Software reliant on this nascent technology, one rooted in the physical laws of nature, could soon revolutionize computing forever. Bear in mind, however, that it took classical computing many decades to go from individually programmed logic gates to the sophisticated cloud-based services of today—and we hope to see quantum computing take that same leap in just a few short years. We think we can get there, but we can’t take this leap alone.
Back in September 2020, we took the bold step of releasing a hardware roadmap showing a clear pathway to over 1,000 qubits, identifying the challenges we anticipate along the way and proposing solutions to these challenges. As scientists, it’s not an easy decision to go public with such a transparent roadmap; we prefer to talk about our achievements, not our plans. However, creating the right conditions for quantum computing to transform the broader distributed computing ecosystem within this decade is a gargantuan effort. We must incorporate the past 50 years of lessons learned from classical computing to accelerate the integration of quantum and classical systems at an astronomically fast pace. Thankfully, this time around we’ll have the help of an open-source community’s power and ability to mobilize developers worldwide, plus cloud-native deployment of quantum workloads, in order to break down any potential barriers and to democratize access to this new technology as quickly as possible.
We know that software development is best done collaboratively, since open-source approaches are based around the understanding that an ecosystem of different human needs drives the best outcomes — and quantum computing is no different. We expect developers to work in each of three key segments laying the groundwork for those working higher up in the stack.
- At the lowest level, quantum kernel developers are creating high-performance quantum circuits with timing and pulse-level controls.
- Quantum algorithm developers rely on these circuits to develop groundbreaking quantum algorithms that might provide an advantage over present-day classical computing solutions.
- Finally, quantum model developers apply these algorithms to real-world use cases in order to develop quantum models for chemistry, physics, biology, machine learning, optimization, or even finance.
” — Ilyas Khan, founder and CEO of Cambridge Quantum Computing
Our Development Roadmap serves to give each developer segment the tools they need to produce the best circuits, algorithms, and models, while maximizing the opportunities for collaboration. We are increasing the variety of circuits and the capacity of our systems to run more circuits more quickly, while developing a platform where quantum developers can work seamlessly in the same integrated cloud-based framework. Workloads with both quantum and classical components will not be constrained by origin or the nature of integration, and the hybrid cloud will allow these workloads to run everywhere that our cloud native systems run today and in the future.
So, how do we plan on getting there?
Kernels. Algorithms. Models.
Today, we’re making crucial updates for quantum kernel developers writing code at the lowest level, for whom we’ve been focusing on developing circuit APIs. This year, we plan to release the Qiskit runtime—an execution environment that increases the capacity to run more circuits at a much faster rate than ever before, and with the capability to store quantum programs so other users can run them as a service. The Qiskit runtime rethinks the classical-quantum workload so that programs will be uploaded and executed on classical hardware located beside quantum hardware, slashing latencies emerging from communication between the user’s computer and the quantum processor.
“As a researcher and start-up founder, I’m more than familiar with the challenges of turning research results into a valuable product. The IBM Quantum developer roadmap will be enormously helpful in shaping our goals to create differentiated values in this ecosystem, and for directing research topics and efforts to achieve the maximum relevance and impact for my lab and collaborators. — Dr. Prineha Narang, founder and CTO at Aliro Quantum and assistant professor at the John A. Paulson School of Engineering and Applied Sciences at Harvard University
These improvements will lead to a 100x speedup in workloads that exploit iterative circuit execution, which will allow our quantum systems to run jobs in just a few hours that, today, can take months.
Not only are we increasing the capacity of our systems, but our system will run a wider variety of circuits, allowing users to tackle problems previously inaccessible to any quantum processors. Updated software interfaces, like our recently announced OpenQASM3 assembly language, plus ongoing technical development will allow quantum kernel developers to run dynamic circuits—those incorporating both classical and quantum instructions that need to be run within the coherence time of the qubits—by 2022. Into 2023 and beyond, we will debut circuit libraries and advanced control systems for manipulating large qubit fabrics, allowing our quantum kernel developers to take full advantage of hardware with a thousand or more qubits.
Our quantum algorithm developers have different concerns than our quantum kernel developers, requiring efficient execution of jobs as they explore potentially groundbreaking new applications of quantum circuits. These users will be looking for new ways to take advantage of the power of quantum computing, without having to think about the intricacies of the hardware on which they’re programming. For these developers, we will continue improving and rolling out new tools to efficiently build and implement quantum circuits, as well as application-specific modules for natural science, optimization, machine learning, and finance to make their exploration easier.
— Will Zeng, founder, Unitary Fund
By 2023, we expect to offer entire families of pre-built runtimes tailored to these domains, callable from a cloud-based API using a variety of common development frameworks. At this point, we think the foundations laid down by quantum kernel and algorithm developers will allow model and enterprise developers to explore quantum computing models on their own without having to think about the quantum physics. Developers will have the freedom to enrich systems built in any cloud-native hybrid runtime, language, and common programming framework, or integrate quantum components simply into any business workflow.
A frictionless quantum development future
Looking to 2025 and beyond, we think that our dream of frictionless quantum computing will become a reality—one where the hardware is no longer a concern to users or developers. By then, we envision that developers across all levels of the quantum computing stack will rely upon on our advanced hardware with a cloud-based API, working seamlessly with high performance computing resources to push the limits of computation overall—and include quantum computation as a natural component of their existing computation pipelines. We hope that by 2030, companies and users are running billions, if not a trillion quantum circuits a day, perhaps without even realizing that they’re doing so.
We hope that our roadmap toward this truly frictionless development framework can deliver the best experience with quantum computers in the world for our users, clients, and our quantum computing community. And we certainly do not plan to take this journey alone. Through continued release of open-source software for the community to explore, create, and develop exciting new quantum applications, we will build the quantum ecosystem of tomorrow, together.
“IBM is unique in their approach of sharing this transparent roadmap. Fostering an open, collaborative, ecosystem of startups, researchers and enterprise developers is core to the Strangeworks mission. We’re excited by the prospect of deploying IBM’s quantum technologies across the Strangeworks family of products allowing our users to explore hybrid cloud runtimes today, and employ 1,000+ qubits within the next few years.” — whurley, founder and CEO of Strangeworks
A big part of our software strategy is to continue to use and create open source tools, eventually converting some into first-class cloud native components. This will allow us to continue scaling and extending our quantum software so that users can take advantage of our architecture while running quantum programs in a secure and reliable way. On the other side, users will be able to install and use some components from our software stack directly in their preferred cloud architectures.
And as we roll out this roadmap, we want to hear from you; If you’re a developer who wants to be able to take advantage of quantum, and are interested in being part of our feedback program to help shape the future of quantum, please sign up here.
Content may have been edited for style and clarity.