We present a framework that utilizes quantum algorithms, an architecture aware quantum noise model and an ideal simulator to benchmark quantum computers. The benchmark metrics highlight the difference between the quantum computer evolution and the simulated noisy and ideal quantum evolutions. We utilize our framework for benchmarking three IBMQ systems. The use of multiple algorithms, including continuous-time ones, as benchmarks stresses the computers in different ways highlighting their behaviour for a diverse set of circuits. The complexity of each quantum circuit affects the efficiency of each quantum computer, with increasing circuit size resulting in more noisy behaviour. Furthermore, the use of both a continuous-time quantum algorithm and the decomposition of its Hamiltonian also allows extracting valuable comparisons regarding the efficiency of the two methods on quantum systems. The results show that our benchmarks provide sufficient and well-rounded information regarding the performance of each quantum computer.
Application-based benchmarks are increasingly used to quantify and compare quantum computers' performance. However, because contemporary quantum computers cannot run utility-scale computations, these benchmarks currently test this hardware's performance on ``small'' problem instances that are not necessarily representative of utility-scale problems. Furthermore, these benchmarks often employ methods that are unscalable, limiting their ability to track progress towards utility-scale applications. In this work, we present a method for creating scalable and efficient benchmarks from any quantum algorithm or application. Our subcircuit volumetric benchmarking (SVB) method runs subcircuits of varied shape that are ``snipped out'' from some target circuit, which could implement a utility-scale algorithm. SVB is scalable and it enables estimating a capability coefficient that concisely summarizes progress towards implementing the target circuit. We demonstrate SVB with experiments on IBM Q systems using a Hamiltonian block-encoding subroutine from quantum chemistry algorithms.
Creating scalable, reliable, and well-motivated benchmarks for quantum computers is challenging: straightforward approaches to benchmarking suffer from exponential scaling, are insensitive to important errors, or use poorly-motivated performance metrics. Furthermore, curated benchmarking suites cannot include every interesting quantum circuit or algorithm, which necessitates a tool that enables the easy creation of new benchmarks. In this work, we introduce a software tool for creating scalable and reliable benchmarks that measure a well-motivated performance metric (process fidelity) from user-chosen quantum circuits and algorithms. Our software, called
scarab, enables the creation of efficient and robust benchmarks even from circuits containing thousands or millions of qubits, by employing efficient fidelity estimation techniques, including mirror circuit fidelity estimation and subcircuit volumetric benchmarking.
scarab provides a simple interface that enables the creation of reliable benchmarks by users who are not experts in the theory of quantum computer benchmarking or noise. We demonstrate the flexibility and power of
scarab by using it to turn existing inefficient benchmarks into efficient benchmarks, to create benchmarks that interrogate hardware and algorithmic trade-offs in Hamiltonian simulation, to quantify the in-situ efficacy of approximate circuit compilation, and to create benchmarks that use subcircuits to measure progress towards executing a circuit of interest.