Qubit Hardware Database

A comprehensive database tracking the evolution and performance of quantum computing hardware across different architectures.

See example plots in "Qubit Database.ipynb" file

Overview

The Qubit Hardware Database is a research project that collects and organizes data on quantum computing hardware from published research papers. The database tracks key performance metrics such as:

• Coherence times (T1, T2)
• Gate fidelities (single and two-qubit)
• Error rates (measurement, state preparation, SPAM)
• Gate operation times
• Qubit counts
• Operating temperatures

This repository provides tools to access, visualize, and analyze trends in quantum computing hardware development across different architectures (superconducting, trapped ion, photonic, etc.) and years.

Data Source

The database is populated from a Google Sheet that contains manually entered data from research papers. Each entry includes:

• DOI (Digital Object Identifier) of the paper
• Hardware type
• Year of publication
• Number of qubits
• Performance metrics
• Summary of the work
• Additional notes

Using the Database

Database Structure

The database contains:

1. A main table with basic information about each paper
2. Normalized tables for specific metrics:
   • T1_Times
   • T2_Times
   • Single_Qubit_Fidelity
   • Two_Qubit_Fidelity
   • Measurement_Error
   • State_Prep_Error
   • SPAM_Error
   • Single_Qubit_Gate_time
   • Two_Qubit_Gate_time

Accessing Data

import sqlite3
import pandas as pd

# Connect to the database
conn = sqlite3.connect('qubit_data.db')

# Simple query to get all data from main table
main_df = pd.read_sql_query("SELECT * FROM main", conn)

# Query specific metrics
t1_data = pd.read_sql_query("SELECT * FROM T1_Times", conn)

# Join metrics with main table information
query = """
SELECT m.DOI, m.Hardware_Type, m.Year, m.Number_of_Qubits,
       t.Value, t.Uncertainty
FROM T1_Times t
JOIN main m ON t.DOI = m.DOI
"""
combined_data = pd.read_sql_query(query, conn)

# Always close the connection when done
conn.close()

Creating Visualizations

Basic Time Series Plots

import matplotlib.pyplot as plt
import seaborn as sns
import sqlite3
import pandas as pd

# Connect to the database
conn = sqlite3.connect('qubit_data.db')

# Load data with joins
def load_table_data(table_name):
    query = f"""
    SELECT m.DOI, m.Hardware_Type, m.Year, m.Number_of_Qubits,
           t.Value, t.Uncertainty
    FROM {table_name} t
    JOIN main m ON t.DOI = m.DOI
    """
    return pd.read_sql_query(query, conn)

# Example: Plot T1 Times vs Year
df = load_table_data('T1_Times')
df['Year'] = pd.to_numeric(df['Year'], errors='coerce')
df = df.dropna(subset=['Year', 'Value'])

plt.figure(figsize=(12, 6))
sns.scatterplot(
    data=df,
    x='Year',
    y='Value',
    hue='Hardware_Type',
    s=50,
    alpha=0.8
)
plt.yscale('log')
plt.title('T1 Coherence Times Over Years by Hardware Type')
plt.xlabel('Year')
plt.ylabel('T1 Time (ms, log scale)')
plt.legend(title='Hardware Type', bbox_to_anchor=(1.05, 1), loc='upper left')
plt.tight_layout()
plt.show()

conn.close()

The repository includes examples of various visualizations you can create, all included in the "Qubit Database.ipynb" file.

Contributing

You can add new datapoints to this Google Sheet. Here is how new data should be entered:

1. For the normalized data columns if there are multiple entries seperate them by a ";" character.
2. If there are any uncertatnties in measurments add ":" character (e.g. value:uncertanty; ...)
3. For the name of the Hardware type try and keep it the same format as the other names in the previous columns

Future Enhancements

Potential improvements for this project include:

1. LLM-Based Query Interface

   • Create a Large Language Model that uses this database to answer specific questions

2. Web Interface

   • Create a web-based dashboard for exploring the data
   • Implement interactive visualizations with filters

3. Automated Data Collection

   • Develop scripts to scrape papers for relevant metrics
   • Implement natural language processing to extract quantum hardware details

4. Extended Database Schema

   • Add more metrics (e.g., quantum volume, circuit depth capabilities)
   • Track more details about architecture (e.g., connectivity, control systems)