• Context
• Choosing the index type
• Choosing how to partition the vectors
• Choosing the encoding
• Conclusion
• Building the index
• Split across database partitions
• Split across inverted lists
• Using the index
• Search performance and statistics
• Read patterns and speed
• Impact on search speed
• Distributed setting
• Conclusion