Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Visual Analytics of Performance of Quantum Computing Systems and Circuit Optimization

Junghoon Chae, Chad A. Steed, Travis S. Humble

TL;DR

Quantum devices suffer from noise and decoherence, e.g., $T_2$, which degrades fidelity and complicates performance analysis. The paper introduces QVis, a visual analytics dashboard that links topology, multi-scale time-series, clustering, similarity, and distribution views with circuit-optimization visualization via the IBM Qiskit transpiler to compare optimization levels. Contributions include multi-view, interactive analysis of qubit performance and device topology; temporal clustering to identify subgroups of qubits with similar behavior; and integrated visualization of circuit optimization effects on depth and gate counts using real-device data from a $127$-qubit IBM Washington dataset over 16 months. The framework improves interpretability of quantum computations and informs algorithm and hardware-aware design for more reliable quantum performance.

Abstract

Driven by potential exponential speedups in business, security, and scientific scenarios, interest in quantum computing is surging. This interest feeds the development of quantum computing hardware, but several challenges arise in optimizing application performance for hardware metrics (e.g., qubit coherence and gate fidelity). In this work, we describe a visual analytics approach for analyzing the performance properties of quantum devices and quantum circuit optimization. Our approach allows users to explore spatial and temporal patterns in quantum device performance data and it computes similarities and variances in key performance metrics. Detailed analysis of the error properties characterizing individual qubits is also supported. We also describe a method for visualizing the optimization of quantum circuits. The resulting visualization tool allows researchers to design more efficient quantum algorithms and applications by increasing the interpretability of quantum computations.

Visual Analytics of Performance of Quantum Computing Systems and Circuit Optimization

TL;DR

Quantum devices suffer from noise and decoherence, e.g., , which degrades fidelity and complicates performance analysis. The paper introduces QVis, a visual analytics dashboard that links topology, multi-scale time-series, clustering, similarity, and distribution views with circuit-optimization visualization via the IBM Qiskit transpiler to compare optimization levels. Contributions include multi-view, interactive analysis of qubit performance and device topology; temporal clustering to identify subgroups of qubits with similar behavior; and integrated visualization of circuit optimization effects on depth and gate counts using real-device data from a -qubit IBM Washington dataset over 16 months. The framework improves interpretability of quantum computations and informs algorithm and hardware-aware design for more reliable quantum performance.

Abstract

Driven by potential exponential speedups in business, security, and scientific scenarios, interest in quantum computing is surging. This interest feeds the development of quantum computing hardware, but several challenges arise in optimizing application performance for hardware metrics (e.g., qubit coherence and gate fidelity). In this work, we describe a visual analytics approach for analyzing the performance properties of quantum devices and quantum circuit optimization. Our approach allows users to explore spatial and temporal patterns in quantum device performance data and it computes similarities and variances in key performance metrics. Detailed analysis of the error properties characterizing individual qubits is also supported. We also describe a method for visualizing the optimization of quantum circuits. The resulting visualization tool allows researchers to design more efficient quantum algorithms and applications by increasing the interpretability of quantum computations.
Paper Structure (14 sections, 7 figures)

This paper contains 14 sections, 7 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Visualization for Quantum Circuits
  4. Quantum State and Entanglement Visualization
  5. Visualization of Quantum Device Performance
  6. Quantum System Performance Data
  7. QVis Dashboard Overview
  8. Visual Analytics of Qubit Performance
  9. Multi-scale Temporal Performance Exploration
  10. Clustering Analysis of Qubits
  11. Similarity Distance View
  12. Interactive Hardware Topology Visualization
  13. Visualizing Quantum Circuit Optimization
  14. Conclusions

Figures (7)

  • Figure 1: The distribution of $T_2$ times observed for qubit 4 of the IBM transmon device Washington for the period 1-Jan-2022 to 30-Apr-2023. The distribution of ${T_2}$ underlies variations in system behavior and fluctuations in computational errors that can be revealed through visual analytics.
  • Figure 2: QVis consists of several visualizations including Topology View (1), Multi-Scale Time Series View (2), Qubit Similarity Distance (3), Clustering View (4), and Metric Distribution View (5). Each view supports different analytics but these visualizations are tightly interconnected to support holistic analysis.
  • Figure 3: Multi-scale Time Series View features three linked temporal visualizations: Focus Heatmap (A), Context Heatmap (B), and Focus Line Panels. Panels (A) and (B) work together as a focus+context visualization using a binned representation of the selected device metric. The focus line panel displays the focus time range metric value as a line chart. Currently, the readout error metric is displayed in this view.
  • Figure 4: Aggregated representation of temporal data points to mitigate a visual over-plotting issue.
  • Figure 5: Clustering View: Each line graph represents one clustering. Users can choose the distance metric and how many clusters they want. The different colors of each cluster title make it possible to distinguish between clusters in other views.
  • ...and 2 more figures