ISC High Performance 2026
Birds of a Feather session and workshop talk at ISC High Performance 2026 on practical quantum-HPC integration, covering hardware, software, and algorithmic co-design for near-term deployment.
Interested in inviting me to speak? Send me an email.
Birds of a Feather session and workshop talk at ISC High Performance 2026 on practical quantum-HPC integration, covering hardware, software, and algorithmic co-design for near-term deployment.
Quantum processors need to become first-class resources inside compute infrastructure. I introduce QRMI as open, vendor-neutral middleware for orchestration, observability, and maintenance.
I moderated a Pasqal Thoughts 2026 panel on what it takes for HPC and quantum to work together, with perspectives from CINECA, IBM Quantum, and GENCI.
Quantum-HPC middleware bottlenecks are mostly operational. I outline QRMI and a second-level scheduler for work inside HPC allocations. This improves runtime control without replacing existing schedulers.
QPUs as first-class HPC resources require software redesign. I share deployment lessons on orchestration, scheduling, and observability in production sites. This guides data-center integration decisions.
Quantum resource management interface design remains fragmented. I compare QRMI with QDMI and discuss where they should interoperate. This helps teams choose integration paths with lower lock-in.
QC-HPC systems need portable execution environments. I present architecture choices for QRMI integration, second-level scheduling, and observability. This improves usability for both developers and operators.
Quantum-HPC integration for materials science is becoming practical. Working-group talks focused on workflows, operations, and deployment constraints. This sharpens priorities for near-term application teams.
Quantum-classical supercomputing depends on operational integration. I map neutral-atom QPU infrastructure, co-processing workflows, and scheduling constraints. This clarifies what production adoption actually requires.
Heterogeneous quantum computing needs shared scheduling semantics. I cover QRMI and Slurm-based resource management across real HPC deployments. This reduces integration friction for hybrid workloads.
Quantum-HPC architecture still lacks stable handoff points. EQS3 discussions focused on abstractions and modality-specific stack choices. Clarifying these boundaries improves interoperability planning.
HPC scheduler integration for neutral-atom QPUs is still immature. Hardware-specific constraints break generic assumptions in early stack designs. Model those constraints early to avoid costly rework.
Invited talk at the Munich Quantum Software Forum 2026, held October 14-15 in Munich. Talk title and abstract will be added once finalized.
Quantum software engineering often fails on product fit. I analyze why useful internal tools stay unused despite technical quality. This improves requirements, architecture choices, and long-term maintenance.
Quantum software engineering needs better language abstractions. I present Qadence as a differentiable interface for digital-analog programs. This shortens iteration loops between model design and execution.
Pulse-level simulation for neutral-atom QPUs is advancing quickly. Tensor-network methods and differentiable emulators expose limits in many advantage claims. Use simulation evidence to guide realistic roadmap choices.
Quantum-HPC ecosystem progress is visible at SC24. Booth talks and partner sessions showed where integration work is concrete versus hype. This helps prioritize engineering work with practical upside.
Neutral-atom quantum computing needs realistic framing. I explain applications, hardware constraints, and commercial timelines to an academic audience. This helps teams plan work against real delivery horizons.