Why Manticore Search Beats Elasticsearch: Up to 15× Faster and Fully Open‑Source

1. Introduction

Manticore Search is a high‑performance, open‑source search engine written in C++. Launched in 2017 as the successor to Sphinx Search, it rewrites most of Sphinx’s code, fixes hundreds of bugs, and adds modern capabilities, making it a lightweight, fast, full‑text database.

Independent testing by Microsoft shows that Manticore’s performance can surpass Elasticsearch, with speed improvements of up to fifteen times in certain scenarios.

2. Advantages

Very fast and cost‑effective; for small datasets it is 182× faster than MySQL (reproducible).

For log analysis it is 29× faster than Elasticsearch (reproducible).

For small datasets it is 15× faster than Elasticsearch (reproducible).

For medium‑size datasets it is 5× faster than Elasticsearch (reproducible).

For large datasets it is 4× faster than Elasticsearch (reproducible).

Data import on a single server can achieve up to 2× higher throughput than Elasticsearch (reproducible).

Modern multithreaded architecture fully utilizes all CPU cores for the fastest response times.

Powerful, fast full‑text search works seamlessly on both small and large data sets.

Row‑level storage is provided for small, medium, and large datasets.

Columnar storage via the Manticore Columnar Library handles datasets that do not fit in memory.

Automatic creation of efficient secondary indexes saves time and effort.

Cost‑optimised query optimiser maximises search performance.

SQL‑based native syntax compatible with the MySQL protocol, allowing use of any MySQL client.

Easy integration through clients for PHP, Python, JavaScript, Java, Elixir, Go, and others.

Provides an HTTP JSON protocol for flexible data and schema management.

Built in C++ for rapid startup, minimal memory usage, and low‑level optimisations.

Real‑time inserts make newly added documents instantly searchable.

Built‑in replication and load‑balancing increase reliability.

Simple data sync from MySQL, PostgreSQL, ODBC, XML, CSV and other sources.

Supports transactions and binlog for safe writes, although it is not fully ACID‑compliant.

Integrated tools and SQL commands simplify backup and recovery.

Used by companies such as Craigslist, Socialgist, PubChem, Rozetka and many others for efficient search and stream filtering.

3. Getting Started with Docker

Docker images are available on Docker Hub: https://hub.docker.com/r/manticoresearch/manticore/

docker run -e EXTRA=1 --name manticore --rm -d manticoresearch/manticore \
  && until docker logs manticore 2>&1 | grep -q "accepting connections"; do sleep 1; done \
  && docker exec -it manticore mysql \
  && docker stop manticore

After the container is running you can create tables, insert data, and perform searches. Example commands:

create table movies (title text, year int) morphology='stem_en' html_strip='1' stopwords='en';

insert into movies (title, year) values
  ('The Seven Samurai', 1954),
  ('Bonnie and Clyde', 1954),
  ('Reservoir Dogs', 1992),
  ('Airplane!', 1980),
  ('Raging Bull', 1980),
  ('Groundhog Day', 1993),
  ('Jurassic Park', 1993),
  ('Ferris Bueller''s Day Off', 1986);

select highlight(), year from movies where match('the dog');
select highlight(), year from movies where match('days') facet year;
select * from movies where match('google');

Full documentation and the open‑source code are hosted on GitHub: https://github.com/manticoresoftware/manticoresearch