QHub

Meet QHub: An integrated data science environment developed by the scientists at Quansight.

QHub is a full-service, managed instance of JupyterHub on compute resources of your choosing (cloud or on-prem). It includes tooling for simple, rapid stand-up and management, and uses Kubernetes for containerization and Dask for seamless distribution of numerical computation. As your team and their scope of work expand, they’ll ultimately outgrow the computational capabilities of a single machine and require a distributed compute infrastructure. QHub delivers that infrastructure quickly, easily, and with a minimum of cost, since it’s built entirely from open source components.

QHub Is ...

Quick to Install

Set up QHub quickly on the cloud provider of your choice. It deploys in 30 minutes or less with minimal DevOps or Kubernetes experience needed.

Easy to Manage

Control your QHub deployment from an easy-to-use, centralized dashboard. Deploy, maintain, and scale with a single source of truth for configuration.

Highly Collaborative

Share your scientific computing resources seamlessly across distributed teams. QHub builds on the powerful collaboration workflows enabled by JupyterHub.

Easily Scalable

Scale your compute infrastructure in tandem with your team and projects. QHub seamlessly integrates Terraform, Kubernetes, Dask, and others.

Cost Effective

Use only the compute resources you need for your current scale. Customize your compute instance options, including GPUs, plus auto-shutdown policies for idle instances.

Secure

Keep your data, analysis, and results protected with QHub’s security features. Keycloak integration provides user authentication including SSO with minimum effort.

Open Source Projects QHub Includes in Its Managed Integration Platform

Explore Quansight's Consulting Areas

Infrastructure, Scaling, & Acceleration

Visualization & Dashboards

Algorithms, AI, & Machine Learning

Packaging & Environment Management

Data Engineering & MLOps

Jupyter Technologies

We use QHub extensively for our own distributed, collaborative scientific projects—we'll help you do the same.