Search at Scale
The architecture of ArangoSearch is designed to handle even complex queries against large, distributed data sets efficiently.
Parallel, Distributed Query Execution
ArangoDB moves your queries where your data resides. Each DBserver in an ArangoDB cluster has its own query engine and executes your query locally and in parallel.
Fast SIMD Compression
Even terabytes of your data can be stored and searched with C++ based ArangoSearch in ArangoDB. By handling compression on a very low level SIMD can be handled efficiently for optimal performance.
ArangoSearch and ArangoDB support distributed indexes so each shard has its own index for local lookups and query execution.
Combine Full-text Search with Graph
Extend your ArangoSearch queries with graph, geo or relational-type aspects for even more fine-grained access to your data. Combine your search with unique features like SmartGraphs or SmartJoins for high performance queries against distributed data.