Amazon Web Services (AWS) and its research partners are making the case that classical data center infrastructure may become a key enabler of quantum computing progress, following a new demonstration of a 97-qubit error-correction simulation on cloud HPC.
In a blog post, researchers from AWS, Quantum Elements, the University of Southern California, and Harvard University describe a “hardware-calibrated digital twin” approach designed to model quantum error behavior while remaining computationally tractable.
Simulating QEC at Experiment-Relevant Scale
The team reports simulating a distance-7 rotated surface code comprising 97 physical qubits – a scale comparable to recent experimental systems – using classical infrastructure on Amazon Elastic Compute Cloud (EC2). According to the post, the simulation models a full syndrome-extraction cycle and completed in approximately one hour on a single EC2 Hpc7a instance using 96 vCPUs. That scale would be infeasible with conventional…