Sonic
Fast, lightweight and schema-less search backend written in Rust -- an Elasticsearch alternative for simple text search that runs on a few MB of RAM.
- Actively maintained
Overview
Sonic is an identifier index rather than a document store: you push text along with an object ID, and searches return matching IDs that you then look up in your own database. Keeping only the index makes it extremely small and fast, with memory use measured in megabytes rather than gigabytes.
It supports search, autocomplete suggestions and typo tolerance, organizes data into collections and buckets, and is controlled over a simple TCP protocol with client libraries in many languages. It was built to power search at the customer-support company Crisp. A good fit when you need quick search or autocomplete over an existing database and a full Elasticsearch cluster would be overkill.
Why it's in this catalog
Catalog rating: ★★★☆☆ (3/5)
Tiny, schema-less search index that runs in a few MB of RAM; a niche but distinctive option far lighter than anything else in the category.
- Mature
- Community-driven
- Categories
- Search
- Languages
- Rust
- License
- MPL-2.0
Related Projects
OpenSearch
Apache-2.0Apache-2.0 licensed distributed search and analytics engine, forked from Elasticsearch to guarantee a fully open-source alternative for full-text search and log analytics.
Actively maintained
- Search
Elasticsearch
AGPL-3.0Distributed search and analytics engine for all types of data, including text, numerical and geospatial.
Actively maintained
- Search
Meilisearch
MITFast, open source search engine that is easy to integrate and delivers relevant results instantly.
Actively maintained
- Search
Quickwit
Apache-2.0Cloud-native search engine purpose-built for observability data, indexing logs and traces directly on object storage for cost-efficient search at scale.
Actively maintained
- Search
Spotted an error? Suggest an edit on GitHub.