Uncluster Your Data Science Using Vaex

Explore how to build snappy dashboards visualizing hundreds of millions of data points and interactively explore hundreds of gigabytes of data on a single machine using Vaex, an out-of-core DataFrame library in Python. Learn about memory mapping, column-based storage, and the compute and expression system that enables Vaex to perform typical data manipulations, filtering, and aggregations on a billion rows in real-time. Discover how this approach can empower your team by removing the DevOps overhead of configuring and maintaining a cluster. Watch a comprehensive demo showcasing Vaex's capabilities, and gain insights into its production use cases, including examples with Dash. Understand how Vaex evolved from an academic project to a consultancy, and explore its potential applications in data science and machine learning workflows.

Intro
Motivation
Vaex
Concepts: Memory mapping
Concepts: Column based storage
Concepts: No memory copies
Concepts: Compute & expression system
Vaex.io: From academic project to consultancy
Demo
In production
In the wild
In production: Dash example
Vaex.io: Consultancy
Summary
Outro