Big data. Real time.
The open big data serving engine: Store, search, rank and organize big data at user serving time.
Get Started Now

Features

Vespa's high-level features.

Search and filter
Combine structured queries and text search to select data.
Advanced ranking
Deploy machine learned models for ranking/recommendation.
Organize and aggregate
Group and aggregate all data matching queries, in real time.
Realtime
UI friendly response times, instant writes at high volume.
Scalable and fast
Optimized for Thousands of queries/sec, billions of documents.
Elastic and fault tolerant
Add, remove and replace nodes while live and without losing data.
Pluggable
Deploy your own Java components to implement custom logic.
Easy to operate
Configure systems with a few lines of text, change anything live.
See more

Architecture

The high level architecture of Vespa in a picture.

Vespa architecture diagram

Feature comparison matrix

It is not always easy to compare apples to oranges, but we will give it a try nonetheless.

Feature Vespa Elastic Search Relational Databases
ACID transactions •••
Optimized for analytics ••• ••
Optimized for serving ••• • ••
Scalable ••• •• •
Easy to operate at scale •• •
Text search ••• •• •
Machine learned ranking ••• •
Middleware logic container •••
Live reconfiguration •••