manticoresoftware / manticoresearch

Easy to use open source fast database for search Manticore Search is an easy-to-use, open-source, and fast database designed for search. It is a great alternative to Elasticsearch. Website • Downloads • Docs • Blog • Courses • Forum • Slack • Telegram (En) • Telegram (Ru) • Twitter • User feedback Introduction **❗Read recent blog post about Manticore vs Elasticsearch❗** What distinguishes it from other solutions is: • It's very fast and therefore more cost-efficient than alternatives, for example Manticore is: • **182x faster** than MySQL for small data (reproducible❗) • **29x faster** than Elasticsearch for log analytics (reproducible❗) • **15x faster** than Elasticsearch for small dataset (reproducible❗) • **5x faster** than Elasticsearch for medium-size data (reproducible❗) • **4x faster** than Elasticsearch for big data (reproducible❗) • **up to 2x faster** max throughput than Elasticsearch's for data ingestion on a single server (reproducible❗) • With its modern multithreading architecture and efficient query parallelization capabilities, Manticore is able to fully utilize all your CPU cores to achieve the quickest response times possible. • The powerful and speedy full-text search works seamlessly with both small and large datasets. • Row-wise storage for small, medium and big size datasets. • For even larger datasets, Manticore offers columnar storage support through the Manticore Columnar Library, capable of handling datasets too big to fit in RAM. • Performant secondary indexes are automatically created using the PGM-index (Piecewise Geometric Model index), which provides efficient mapping between indexed keys and their memory locations. • The cost-based query optimizer uses statistical data about indexed data to evaluate and determine the most efficient execution plan. • Manticore is SQL-first, utilizing SQL as its native syntax, and offers compatibility with the MySQL protocol, allowing you to use your preferred MySQL client. • With clients available in PHP, Python, Python asyncio JavaScript, Typescript, Java, Elixir, Go, and Rust, integration with Manticore Search becomes easy. • Manticore also provides a programmatic HTTP JSON protocol for more versatile data and schema management, with Elasticsearch-compatible writes support. • Built in C++, Manticore Search starts quickly and uses minimal RAM (around 40MB RSS for an empty instance), with low-level optimizations contributing to its impressive performance. • With real-time inserts, newly added or updated documents are immediately accessible. • Interactive courses are available through Interactive courses to make learning a breeze. • Manticore also boasts built-in virtually synchronous multi-master replication using the Galera library and load balancing capabilities. • Data can be synced from sources such as MySQL, PostgreSQL, ODBC, xml, and csv with ease. • While not fully ACID-compliant, Manticore supports isolated transactions and binary logging for safe writes. • Effortless data backup and recovery with built-in tools like manticore-backup and SQL BACKUP command Craigslist, Socialgist, PubChem, Rozetka and many others use Manticore for efficient searching and stream filtering. Manticore Search was forked from Sphinx 2.3.2 in 2017. More features • Full-text search and relevance: • Over 20 full-text operators and over 20 ranking factors • Custom ranking • Other search capabilities: • Rich filtering functionality • Fuzzy search • Faceted search • Geo-spatial search • Vector search • Joining tables • Spelling correction • Autocomplete • A wide range of functions for filtering and data manipulation • Natural language processing (NLP): • Stemming • Lemmatization • Stopwords • Synonyms • Wordforms • Advanced tokenization at character and word level • Proper Chinese segmentation • Text highlighting • Stream filtering: • using a "percolate" table • or the Kafka integration • High-availability: • Data can be distributed across servers and data-centers • Synchronous replication • Built-in load balancing • Security: • https support • read-only mode • Data safety: • manticore-backup tool and SQL command BACKUP to back up and restore your data • Data storages: • row-wise - requires more RAM, provides faster performance • columnar - requires less RAM, still provides decent performance, but lower than the row-wise storage for some kinds of queries • docstore - doesn't require RAM at all, but allows only fetching original value, not sorting/grouping/filtering • Performance optimizations: • Secondary indexes • Cost-based optimizer determines the most efficient execution plan of a search query • Data types: • full-text field - inverted index • int, bigint and float numeric fields in row-wise and columnar fashion • multi-value attributes (array) • string and JSON • on-disk "stored" for key-value purpose • Integrations: • Sync from MySQL and PostgreSQL • Sync from XML • Sync from CSV • Sync from ODBC • Sync from MS SQL • Sync from Kafka • With MySQL as a storage engine • With MySQL via FEDERATED engine • ProxySQL • Apache Superset • Grafana • Fluentbit • Kibana (Demo) • Logstash/Filebeat • Vector.dev • Mysqldump • Manticore Columnar Library Installation Docker Docker image is available on Docker Hub. To experiment with Manticore Search in Docker just run: You can then: create a table, add data and run searches. For example: Note that upon exiting the MySQL client, the Manticore container will be stopped and removed, resulting in no saved data, so **use this way only for testing / sandboxing purposes**. Read the full instruction for the docker image for more details including our recommendations on running it in production. Packages Ubuntu, Debian, Centos, Windows and MacOS packages are here. YUM repo for RHEL/Centos/Amazon/Oracle Linux APT repo for Ubuntu/Debian/Mint Homebrew on MacOS Windows See instruction here. Clouds • Elestio • Hosting Ukraine Documentation and community sites • Documentation • Interactiv…