PROVIDER Β· Mature Java full-text search platform
Apache Solr
Apache Solr is the mature open source full-text search platform on Lucene whose faceting, highlighting, and query parsers were refined over nearly two decades, fitting existing enterprise search and classic site retrieval.
Pros
Cons
Product Information
Detailed Notes
Overview
Apache Solr was created by Yonik Seeley at CNET in 2004 and became an Apache top-level project in 2006, built on Lucene as the earliest widely adopted enterprise open source search engine, long used for site search, ecommerce retrieval, and document management.
For Backend & Database work the advantage is maturity and control, since faceting, highlighting, spell check, synonyms, and rich query parsers were refined over nearly two decades, Chinese tokenisation supports IK and SmartCN, and SolrCloud uses ZooKeeper for distributed coordination with shards and replicas.
The gap is innovation pace, because vector search and learned ranking trail Elasticsearch and Vespa, cloud-native and container ecosystems are less active than OpenSearch, community momentum declined in recent years, and new projects favour Elasticsearch or OpenSearch with Solr appearing mostly in existing systems.
Core Strengths
- Built on Lucene with nearly two decades of refinement in full text, faceting, and highlighting
- SolrCloud shards with replicas coordinated through ZooKeeper
- Chinese tokenisation supports IK, SmartCN, and other schemes with flexible configuration
- Apache 2.0 licensing with no restrictions
Product Ecosystem
Core and queries
Lucene indexes with rich query parsers covering boolean, phrase, fuzzy, and function queries.
Distributed deployment
SolrCloud manages state via ZooKeeper with collection sharding, replicas, and automatic failover.
Chinese and extensions
Analyser chains in the schema plus synonym files, stop words, and custom request handlers.
Drawbacks
- Vector search and learned ranking trail Elasticsearch and Vespa
- Community momentum declined with new projects favouring Elasticsearch or OpenSearch
- The ZooKeeper dependency complicates deployment and cloud-native support lags OpenSearch
- Chinese tokenisation and relevance tuning demand substantial manual configuration experience
Use Cases
- Maintaining existing enterprise search (β β β β ) - Mature and stable
- Site full text with faceting (β β β β ) - Solid classic capability
- Document and archive retrieval (β β β β ) - Rich query parsers
- Greenfield vector search (β β ) - Weaker capability
Pricing
| Plan | Price | Scope |
|---|---|---|
| Apache Solr open source | $0 | Apache 2.0, self-managed |
| Managed Solr | Per instance size | Hosted with backup |
| Commercial distribution and support | Contract quoted | Third-party vendor support |
The open source edition is free with cost in servers and staff, since SolrCloud needs ZooKeeper and multi-node planning, and where a team already knows Solr, continuing costs less than migrating to Elasticsearch.
FAQ
-
Solr or Elasticsearch? Solr for maintaining existing deployments, Elasticsearch or OpenSearch for new projects needing vector search and an active ecosystem.
-
How is Chinese tokenisation configured? Commonly with IK or SmartCN analysers plus synonyms and custom dictionaries, tuned iteratively against business corpora.
-
Is Solr being retired? The project still releases updates while community momentum declines, so existing systems continue while new projects weigh ecosystem and hiring.