Noise-Adaptive Compiler Mappings for Noisy Intermediate-Scale Quantum Computers (1901.11054v1)

Abstract: A massive gap exists between current quantum computing (QC) prototypes, and the size and scale required for many proposed QC algorithms. Current QC implementations are prone to noise and variability which affect their reliability, and yet with less than 80 quantum bits (qubits) total, they are too resource-constrained to implement error correction. The term Noisy Intermediate-Scale Quantum (NISQ) refers to these current and near-term systems of 1000 qubits or less. Given NISQ's severe resource constraints, low reliability, and high variability in physical characteristics such as coherence time or error rates, it is of pressing importance to map computations onto them in ways that use resources efficiently and maximize the likelihood of successful runs. This paper proposes and evaluates backend compiler approaches to map and optimize high-level QC programs to execute with high reliability on NISQ systems with diverse hardware characteristics. Our techniques all start from an LLVM intermediate representation of the quantum program (such as would be generated from high-level QC languages like Scaffold) and generate QC executables runnable on the IBM Q public QC machine. We then use this framework to implement and evaluate several optimal and heuristic mapping methods. These methods vary in how they account for the availability of dynamic machine calibration data, the relative importance of various noise parameters, the different possible routing strategies, and the relative importance of compile-time scalability versus runtime success. Using real-system measurements, we show that fine grained spatial and temporal variations in hardware parameters can be exploited to obtain an average $2.9$x (and up to $18$x) improvement in program success rate over the industry standard IBM Qiskit compiler.

• The paper presents a compiler framework that uses daily calibration data to optimize qubit mapping and operation scheduling for NISQ systems.
• It compares SMT-based optimization with heuristic approaches, demonstrating an average 2.9x improvement in program success rates.
• These results highlight the practical impact of noise-adaptive compilation on enhancing the reliability and performance of near-term quantum computers.

Noise-Adaptive Compiler Mappings for Noisy Intermediate-Scale Quantum Computers

This paper addresses a critical challenge in the field of quantum computing: effectively mapping and optimizing quantum programs for execution on Noisy Intermediate-Scale Quantum (NISQ) computers. NISQ systems, characterized by their limited number of qubits (up to 1000) and susceptibility to noise and variability, pose significant challenges due to their inability to leverage error correction codes. The authors propose a backend compiler framework that adapts to the dynamic characteristics of NISQ hardware to improve execution reliability and success rates significantly.

The framework begins by translating quantum programs represented in a high-level language like Scaffold into an LLVM intermediate representation (IR). The compiler then employs a combination of heuristic and optimization-based methods to generate quantum executable code, which can be run on IBM Q quantum machines. Key to the approach is the utilization of daily calibration data from the quantum hardware, which allows for smart decision-making regarding qubit mapping and operation scheduling that takes into account coherence times, error rates, and other parameters crucial to execution reliability.

Several main components underpin the proposed compiler framework:

• Qubit Mapping: The mapping strategy aims to optimize the placement of logical program qubits onto physical hardware qubits, a critical step that minimizes the need for qubit movement and reduces exposure to high-error qubits.
• Operation Scheduling: Efficient operation scheduling is employed to ensure that program operations are executed within the coherence time windows of targeted qubits.
• Qubit Routing: An efficient routing algorithm is implemented to perform necessary SWAP operations, taking into account error rates associated with CNOT gates and the layout of the quantum device.

Two approaches are explored for the compiler mappings:

1. Optimization-Based Compilation: This utilizes an SMT solver to formulate and solve a constrained optimization problem. The optimization aims to maximize program reliability through intelligent qubit assignments and operations scheduling. This involves determining the highest-reliability paths for qubit interactions based on real-time calibration data.
2. Heuristic-Based Compilation: Greedy heuristic approaches provide a faster, albeit potentially less accurate mapping, focusing on minimizing qubit movement and optimizing paths for frequent qubit interactions.

Results presented in the paper demonstrate substantial improvements in program success rates when using the proposed compiler compared to standard tools like IBM's Qiskit. For the evaluations conducted, improved program success rates present an average increase of 2.9x and execution time improvement by 2.7x, indicating that calibration-aware compilation effectively exploits the available hardware's capabilities.

The paper further discusses the implications of these advancements for future NISQ systems. As quantum hardware continues to evolve, the methods demonstrated here are critical for steering quantum computing toward practical applications and accelerating the achievement of quantum supremacy. The robust analysis of calibration data and error tendencies provides a roadmap for the near-term future of quantum hardware, helping to identify architectural areas where error rates could impact performance most significantly. This adaptation framework is crucial for advancing the field, providing a template for future compilers targeting even larger quantum systems.

In conclusion, the authors present a compelling case for the necessity of noise-adaptive compiler techniques in maximizing the utility and reliability of NISQ systems. They provide a foundational step toward more efficient quantum programming practices that could catalyze breakthrough applications in quantum chemistry, cryptography, and beyond. Future work may explore expanding these techniques to other quantum computing platforms and incorporating additional hardware constraints to further optimize performance for large-scale quantum computations.