Quantum Benchmark

Quantum Benchmark
Name	Quantum Benchmark Inc.
Industry	Quantum computing, Software
Founded	2017
Founders	Joseph Emerson, Joel Wallman
Headquarters	Kitchener, Ontario, Canada
Key people	Joseph Emerson (CEO), Joel Wallman (CTO)
Products	True-Q software, Randomized Benchmarking tools
Website	https://www.quantumbenchmark.com

Quantum Benchmark. It is a specialized field and a commercial enterprise focused on developing software and methodologies to characterize, validate, and improve the performance of quantum computing hardware. The core purpose is to provide accurate, scalable metrics that quantify the capabilities and error rates of quantum processors, enabling hardware developers, algorithm designers, and end-users to make informed decisions. This discipline is critical for advancing the technology from laboratory experiments toward practical, fault-tolerant applications, distinguishing between fundamental hardware limitations and correctable noise.

Definition and Purpose

Quantum benchmarking refers to the suite of techniques used to assess the performance of quantum bits (qubits) and the quantum gates that manipulate them. Its primary purpose is to measure key parameters such as gate fidelity, coherence time, and overall processor reliability in a way that is scalable and resistant to systematic errors. This process is essential for guiding the engineering efforts of companies like IBM, Google Quantum AI, and Rigetti Computing, providing a standardized means of tracking progress toward quantum advantage. By establishing a clear performance baseline, it helps allocate resources effectively within the National Quantum Initiative and similar global research programs.

Types of Quantum Benchmarks

Several distinct benchmarking protocols have been developed, each targeting different aspects of quantum processor performance. Randomized Benchmarking and its extension, Clifford Group Randomized Benchmarking, are widely used to estimate average gate fidelities by applying random sequences of gates. Gate Set Tomography provides a more complete, self-consistent picture of all gate operations but at a higher computational cost. For assessing a processor's capability to run specific algorithms, application-oriented benchmarks like the Quantum Volume metric, pioneered by IBM, or cross-entropy benchmarking, famously used in Google's quantum supremacy experiment, are employed. Other types include cycle benchmarking and direct fidelity estimation.

Key Metrics and Performance Indicators

The most critical metrics derived from quantum benchmarks include the average gate fidelity, which quantifies the accuracy of quantum logic operations, and the error rate per gate. Coherence times, specifically T1 and T2 times, measure how long quantum information persists before being lost to decoherence. Quantum Volume is a holistic single-number metric that incorporates qubit count, connectivity, and gate fidelity to indicate a processor's general capability. For error correction research, metrics like the logical error rate and threshold theorem conditions are paramount. These indicators are routinely reported by institutions like NASA's Quantum Artificial Intelligence Laboratory and Intel.

Standardized Benchmarking Protocols

The push for standardization is led by consortia and national bodies to ensure comparability across different hardware platforms. The IEEE Standards Association has working groups, such as the IEEE P7130 group, focused on quantum computing definitions and performance metrics. Organizations like NIST play a key role in developing and validating reference protocols. Standardized protocols define everything from the calibration procedures and sequence design to the data analysis methods, ensuring that results from a superconducting qubit processor at MIT can be meaningfully compared with those from an ion trap system at University of Innsbruck.

Applications and Industry Impact

Effective benchmarking directly impacts the entire quantum technology stack. For hardware developers like Honeywell and Alpine Quantum Technologies, it identifies specific error mechanisms to target for improvement. For software companies such as Zapata Computing and QC Ware, it informs the development of robust algorithms resilient to characterized noise. In the financial sector, firms like Goldman Sachs and JPMorgan Chase use performance data to evaluate the timeline for quantum algorithms in portfolio optimization. The pharmaceutical industry, including partners of Biogen, relies on benchmarks to assess the feasibility of quantum chemistry simulations for drug discovery.

Challenges and Limitations

Significant challenges remain in quantum benchmarking. A primary issue is contextuality, where the error rate of a gate can depend on the surrounding circuit, making simple metrics insufficient. The resource overhead of comprehensive protocols like Gate Set Tomography can be prohibitive for large processors. There is also the problem of non-Markovian noise, which violates the assumptions of many standard techniques. Furthermore, as systems scale, new phenomena like crosstalk between qubits become dominant and are difficult to characterize. These limitations are active research areas within groups at the University of Sydney, Caltech, and QuTech.

Category:Quantum computing Category:Quantum information science Category:Technology companies of Canada