Magento Search Types for Better Performance

With the new Magento Enterprise Edition (EE) and Community Edition you can now configure any of the following as the catalog search engine:

Magento Community:

- Like and Fulltext search type is available as default option
- Combine (Fulltext and Like) search type also available as default method
- Sphinx search platform, which requires third party installation and server configuration
Magento Enterprise:

- Like and Fulltext search type is available as default option
- Combine (Fulltext and Like) search type also available as default method
- Sphinx search platform, which requires third party installation and server configuration
- Apache Solr search platform, which is default for Magento Enterprise, but requires Tomcat web server installation.

In order to install these default search options select the Catalog option in the Configuration tab in the System menu. Then select Catalog search. Generally Like is selected as default search mode. But it is better to activate Fulltext mode, as like makes the resource heavy. After selecting search mode do refresh the Catalog Search index in the Index Management attribute.

MySQL fulltext search (FTS)

It is the most well known and commonly used search as it has a simple setup and requires just some small changes in query. But it only provides basic options, like stop words and minimum word length. Also it doesn’t have a customizable ranking system. Another problem is since MySQL is a database, thus the main purpose is to store data, lacking the efficiency in searching. Fulltext search causes the process to down.

Sphinx Search

It is an open source, easy to integrate search platform. A single Sphinx server has the power to replace several MySQL FTS servers. It not only consumes less memory, but is also much faster than MySQL as it can serve more queries in a second. The cost of implementing as compared to the speed it offers is manageable. Sphinx has the efficiency to utilize server resources based on the availability.


Sphinx supports a part of SQL which offers standard querying. It has all the search functionalists used by the standard SQL syntax along with Sphinx extensions and Sphinx API.


SphinxSE is the MySQL storage engine which can be easily compiled into MySQL server using pluggable architecture. It is a built-in client and does not store any data. It allows MySQL server to run search queries, and obtain required results. One can easily port MySQL FTS applications to Sphinx.

Apache Solr

Solr is the latest and lightening fast open source enterprise search platform offered by Apache Lucene. Its main features include powerful full-text search, faceted search, dynamic clustering, rich document handling, database integration, and Geo spatial search. Solr is highly reliable and offers distributed indexing and load-balanced querying. It also has the ability of automatic recovery as the configuration is centralized. It is now commonly used by some of the world's largest internet sites.

Solr is a standalone fulltext search server. It uses the Lucene Java search library for full-text indexing and search, and can be used from any programming language. The external configuration makes the connectivity with any type of application without Java coding possible as it has extensive plugin architecture.

Solr Features

  • It is optimized to handle high volume of web traffic.
  • It is based on XML, JSON and HTTP.
  • It is linearly scalable, has auto index replication, auto fail over and recovery options.
  • Is adaptable with XML configuration.
  • CopyField functionality allows indexing a single field multiple ways, or combining multiple fields into a single searchable field
  • It has got advanced DisMax query parser for high relevancy results from user-entered queries.
  • It has got centralized Apache ZooKeeper based configuration.

When using Solr or Sphinx for search, in case of incorrect search queries, it automatically takes the correct spellings and display results, this was not in the case of MySQL, which offers no results in this case.

Since you marketers have heavily loaded store and have to handle hundreds of search requests per second, use of Sphinx or Solr search platforms will be the best.

But remember that this will require server reconfiguration of your search server.

Never the less this is not going to stop anyone from using these lightning-fast search tools...

[We are Connecting Dots, a full service eCommerce solutions company. At Connecting Dots, we help companies build great online businesses, we are Magento implementation partners and build great eCommerce portals]