Architectural Foundations for Checkpointing and Restoration in Quantum HPC Systems

TL;DR

Quantum HPC resilience is challenged by the no-cloning and measurement-collapse of quantum states, which breaks traditional state checkpointing. The paper proposes checkpointing at the level of algorithmic and control-flow state, enabled by dynamic quantum circuits that support mid-circuit measurements and classical feedforward, with restoration achieved by re-executing circuits guided by stored classical state. It introduces a layered architecture (runtime/control, program with dynamic circuits, and classical storage) and a taxonomy of checkpoints (classicalized, algorithmic, and logical) along with a coordinating runtime for checkpointing, storage, and restoration. The approach targets near-term feasibility using existing hardware capabilities and HPC storage, offering incremental deployment and a clear pathway toward logical-qubit checkpointing in fault-tolerant regimes. Overall, this framework enables restartable, resilient quantum workflows that align with current HPC ecosystems and iterative quantum algorithms, with practical impact on robustness and scalability.

Abstract

In this work, we explore the design of the checkpointing and restoration for quantum HPC that leverages dynamic circuit technology to enable restartable and resilient quantum execution. Rather than attempting to checkpoint quantum states, our approach redefines checkpointing as a control flow and algorithmic state problem. By exploiting mid-circuit measurements, classical feed forward, and conditional execution supported by dynamic circuits, we capture sufficient program state to allow correct restoration of quantum workflows after interruption or failure. This design aligns naturally with iterative and staged quantum algorithms such as variational eigensolvers, quantum approximate optimization, and time-stepping methods commonly used in quantum simulation and scientific computing.