What is it about?
Quantum computers are powerful, but real ones are small and fragile. One idea is to split up a quantum computation into parts, and run each piece on different quantum devices (kind of like task-sharing) but coordinating them so all the pieces together produce the correct result. The tricky part is that different quantum devices (or “modules”) might have different performance, connect differently, or need different ways of communicating with each other. This work proposes a way to model how these “pieces” depend on each other, when and where quantum operations and communications need to happen, using something the authors call Hybrid Dependency Hypergraphs (HDHs). These hypergraphs are diagrams capturing time and data dependencies for both quantum and classical communication. They allow us to make decisions about when to send quantum information between devices, when to wait, or when to do classical coordination based on the computation at hand. By using the HDH model, one can reason more clearly about the costs (time, communication overhead, synchronization) of distributed quantum computing, and make scheduling decisions, particularly when the system involves different types of quantum hardware that might behave differently. Essentially, the paper shows a way to plan out distributed quantum jobs in advance so that running them across different quantum computers is more efficient, or at least well understood in terms of cost and communication tradeoffs.
Featured Image
Photo by Farai Gandiya on Unsplash
Why is it important?
As quantum computers evolve, it’s unlikely we’ll get one giant, error-corrected machine anytime soon. A more realistic path is to connect many smaller modules (often built with different technologies and reliabilities) into a network that works together on big tasks. The challenge is that splitting work across devices comes with overhead: coordination, communication, and waiting. If not planned well, those costs can outweigh the benefits. Previous work often assumed all modules were identical or that communication was easy. Our approach takes a broader, more realistic view. With Hybrid Dependency Hypergraphs (HDHs), we can represent not just the hardware differences between modules, but also the different computational models of quantum computing, from the familiar circuit model to alternatives like measurement-based quantum computing. HDHs unify quantum and classical dependencies in a single framework, making it possible to see where bottlenecks or expensive communications arise, and to schedule tasks more effectively across a truly diverse system. This is timely because experiments already show that quantum modules can be linked by entanglement and teleportation, and photonic quantum computers in particular already demonstrate distributed execution, but they don’t fit neatly into the standard circuit model that most research assumes. As distributed quantum computing moves from concept to practice, frameworks like HDHs help bridge the gap, capturing both the variety of hardware and the variety of models in which quantum computing is expressed.
Perspectives
From a personal perspective, working on this poster has highlighted for me the critical importance of planning and modelling in distributed quantum computing, not just the quantum algorithms themselves, but the logistics of how different quantum modules talk to one another, how timing and data dependencies propagate across a network, and how classical coordination interacts with quantum operations. It’s tempting, when thinking about quantum computing, to focus only on the “quantum magic”: superposition, entanglement, clever algorithms. But in real systems, much of the challenge comes from how to coordinate quantum operations across devices, especially when those devices are physically separated or built with different technologies. If communication is slow, or unreliable, or expensive, then the overhead from coordinating the distributed pieces might outweigh the benefit of splitting the task and running it in parallel. The HDH abstraction makes these costs explicit and provides a structured way to analyse them for any quantum computing model. Moreover, thinking concretely in terms of HDHs has forced me to confront questions like: When will I need to send entangled qubits over a network? When can I delay communication? When do I need to “store” quantum information, and for how long? These are practical questions but ones that aren’t always obvious when designing quantum algorithms in isolation. I see this work as a step toward bridging the gap between high-level quantum algorithm design and the messy reality of running quantum computing over real networks with real hardware constraints. Ultimately, I hope that by making the costs and dependencies more explicit, people who design distributed quantum systems can make better choices and avoid hidden coordination bottlenecks.
Maria Gragera Garces
University of Edinburgh
Read the Original
This page is a summary of: Distributed Quantum Computing Across Heterogeneous Hardware with Hybrid Dependency Hypergraphs, September 2025, ACM (Association for Computing Machinery),
DOI: 10.1145/3744969.3748417.
You can read the full text:
Contributors
The following have contributed to this page
