NVIDIA Joins the QIR Alliance as the Effort Enters Year Two

Published

steering-committeeorganizationworkstreamsrecap

NVIDIA joins the QIR Alliance steering committee and recap of year one

After its formation a year ago under the Linux Foundation’s Joint Development Foundation for the development of Open Standards, the QIR Alliance is welcoming NVIDIA as the newest addition to the steering committee. NVIDIA is joining Microsoft, Oak Ridge National Laboratory, Quantinuum, Quantum Circuits Inc., and Rigetti Computing (in alphabetical order) in their effort to develop specifications to facilitate integration of quantum programming frameworks with current and future heterogenous quantum processors.

Following last year’s initial demonstrations at Q2B (recording available here), the QIR Alliance is continuing and expanding its investment into enabling multi-processor systems capable of leveraging both classical and quantum resources.

“Quantum computing has the potential to help solve some of the world’s most important problems. Realizing its promise is going to require optimizing the interface between quantum and classical computing to make quantum computing accessible to domain scientists and enable developers to make use of the best tools from both the quantum and classical worlds. NVIDIA is joining the steering committee for the QIR Alliance because it shares the goal of facilitating interoperability throughout the quantum ecosystem and providing a unified platform to leverage today’s limited quantum processing capabilities as well as more powerful systems of the future.” said Alex McCaskey, Senior Quantum Computing Software Architect at NVIDIA.

To that effect, NVIDIA recently introduced QODA to enable programming of quantum processing units (QPUs), GPUs, and CPUs in one system. The programming framework leverages QIR to connect to various quantum backends, like a range of other efforts that were presented in the bi-weekly community calls organized by the QIR Alliance and hosted on the Unitary Fund Discord.

“QIR makes Rigetti QCS more powerful and accessible to our partners and customers, allowing them to easily use the programming language and the tools they prefer. We look forward to supporting more advanced programs and runtimes for QIR as the specification and the community grow.” said Kalan Snyder, Director of Software Engineering, Rigetti.

To accelerate integration efforts across front- and backend providers, the QIR Alliance has recently concluded its first major workstream to define a formal specification for a Base Profile that captures the common denominator supported across all platforms. The workstream consisted of 16 people from 11 different companies and organizations (IBM, Lawrence Berkeley National Laboratory, Microsoft, NVIDIA, Oxford Quantum Circuits, Pasqal, QFrance, Quantinuum, Quantum Circuits Inc., Quantum Machines, Rigetti Computing).

“QIR is becoming a lingua franca for quantum computing just like LLVM is for the classical. Building on LLVM basis, QIR reduces the amount of effort quantum hardware and software providers spend on ensuring interoperability while allowing end users to program in a language of their choice. Quantinuum intends to support native QIR quantum job execution on the H-series hardware soon after the Base Profile is released including state of the art quantum optimization passes provided by TKET. We are excited about the future of QIR-enabled quantum ecosystem.” said Alexander Chernoguzov, Chief Engineer, Quantinuum.

A subsequent workstream to define a more advanced profile that includes support to express adaptive programs is scheduled to launch later this year. To facilitate service integration of hardware backends as well as simulation tools for research and development, an ongoing workstream proceeding in parallel takes first steps towards formalizing the interaction and data exchange between a runtime environment and various accelerators.

“For Quantum Circuits Inc., QIR provides a flexible and comprehensive solution for the representation of hybrid quantum-classical applications that can be executed on our quantum simulators and hardware backends. The ability of QIR to capture all aspects of hybrid computation, including classical computation embedded within a quantum program, allows us to expose our unique hardware capabilities to users via multiple quantum programming language environments.” said Andrei Petrenko, Head of Product, QCI.

More information about ongoing workstreams can be found on the GitHub org page. In addition to facilitating discussions between industry leaders in the form of workstreams and community forums, the QIR Alliance also promotes a dialogue with the research community. Research to accelerate the development of sustainable and scalable quantum computing architectures is a vital part of the effort.

“ORNL is developing new tools to support quantum computing research, including compilation workflows for a diversity of quantum hardware systems. Open development of QIR provides a clear focal point to advance these goals and encourage the best research.” said Travis Humble, Deputy Director of Energy’s Oak Ridge National Laboratory, Oak Ridge National Laboratory.

A similar sentiment is also shared by Wim van Dam, who recently gave the keynote on “Azure Quantum and the Road to Scalable Quantum Computing” at the IEEE Quantum Week 2022.

“We support openness, and this extends to the QIR Alliance and Microsoft’s role as a steering committee member. Azure Quantum leverages QIR to unlock new hybrid quantum computing capabilities – specifically, integrated hybrid quantum computing – and deliver our recently announced quantum resource estimator. Both technical innovations accrue to Azure Quantum’s aim to enable researchers to scale their quantum solutions.” said Wim van Dam, Microsoft.

Also part of the Quantum Week was the second edition of the workshop on Progress and Challenges in Quantum Intermediate Representations. The workshop spawned fruitful discussions across three sessions covering cross platform programming approaches, code transformations and analysis, and interfacing with physical architectures.

“We are excited about the progress we have achieved and are proud of how the effort has grown over the past year. We hope that our work provides a foundation and venue for exchange and collaboration across the industry, research, and the open-source community. We are grateful to be working with a diverse set of thought leaders, innovators, and contributors, and are looking forward to continuing our work as we enter year two of this effort. “ said Bettina Heim, Chair of the QIR Alliance.

For more information about the QIR Alliance, its organization, and how to contribute, visit qir-alliance.org.

Send us updates!

Got some news or QIR related post you want to share? Make a pull request on the repo for this site, or send an email to qiralliance@mail.com.