From the course: Fundamentals of AI Engineering: Principles and Practical Applications
Join today to access over 24,300 courses taught by industry experts.
Scaling strategies (approximate nearest neighbor, or ANN)
From the course: Fundamentals of AI Engineering: Principles and Practical Applications
Scaling strategies (approximate nearest neighbor, or ANN)
- [Instructor] Welcome back, everyone. Today we're going to tackle one of the most important challenges that AI engineers face in production: how to scale their vector databases. To get started, open up chapter five and navigate to 05_04.ipymb. As always, make sure your VN in the upper right-hand corner is selected to the .vn virtual environment. As your AI applications grow, you'll likely move from handling tens, to thousands, to millions, to potentially even billions of vectors at scale. The techniques that we'll cover today are essential for making that transition successfully. Before we dive in, let's establish three primary factors we need to consider when scaling vector databases. First, speed versus accuracy. Understanding when to prioritize one, really, versus the other. Second are resource limitations. How do we work with the memory, CPU, and storage constraints that we have in our operating environment? Third is horizontal scaling. How do we redistribute the workload across…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.