Any examples to use Elastic Search as Graph Database?
It’s important to understand how Graph databases and the likes of elasticsearch differ in how they are optimised.
True graph databases optimise for reads using “index free adjacency” meaning that links between things behave like direct pointers with super cheap O(1) lookup costs. This allows for very fast traversal of many links. Nothing comes for free though so the downside is the index lookups are shifted to index-time with a tax on every write to turn real-world handles like email addresses or bank account numbers into the equivalent internal pointers. Also, to keep traversal costs low, graph dbs advocate keeping all your data on one (often large) machine.
Elasticsearch uses an index for query-time lookups and is designed for storing content on multiple machines so generally has different objectives in mind. That said, not all graph-like reads require the optimisations (and restrictions) a graph database brings.
As usual, “it depends”
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.