Graph Algorithms and Vectors with Neptune Analytics | S02 EP08 | Lets Talk About Data

In this show we dive deep in Graph databases and Amazon Neptune. We discuss about the use cases where customer benefit from using a graph database. In the second half of the show we have a detailed demo to setup and use graph databases

Tony Mullen
Amazon Employee
Published Feb 29, 2024
Last Modified Apr 10, 2024
This show covers an overview of Amazon Neptune and the different components like Neptune Database, Neptune ML, and the new Neptune Analytics. Kevin explains that Neptune Database is more transactional and persistent storage, while Neptune Analytics is optimized for faster analytical queries with the graph fully in memory.
Key highlights:
  • Neptune Database is transactional, persistent SSD storage
  • Neptune ML connects Neptune DB to SageMaker for ML
  • Neptune Analytics is a new service launched at re:Invent for analytical graph workloads, with the graph fully in memory for faster performance
  • Easy to get started with Neptune notebooks and sample data sets
  • Use cases like clustering, community detection for large social networks
  • Supports vector similarity search and embedding directly on the graph
  • Can load data from Neptune DB snapshots or bulk files into Neptune Analytics
  • Output can be exported or written back to Neptune DB
  • Common combination is using Neptune DB for persistence plus Analytics for temporary analytical jobs
The demos show loading sample data sets into Neptune Analytics, running algorithms like pathfinding and breadth-first search, comparing performance to native graph queries, leveraging vector embeddings, and more.
Check out the recording here:
Loading...

Hosts of the show 🎤

Tony Mullen - Senior RDS Specialist Solutions Architect @ AWS

Guests 🎤

Kevin Phillips - Graph Specialist Solutions Architect @ AWS

Links from today's episode

 

Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.

Comments