Indexing overview

Indexing is how Magento transforms data such as products and categories, to improve the performance of your storefront. As data changes, the transformed data must be updated or reindexed. Magento has a very sophisticated architecture that stores lots of merchant data (including catalog data, prices, users, and stores) in many database tables. To optimize storefront performance, Magento accumulates data into special tables using indexers.

For example, if you change the price of an item from $4.99 to $3.99. Magento must reindex the price change to display it on your storefront.

Without indexing, Magento would have to calculate the price of every product on the fly, taking into account shopping cart price rules, bundle pricing, discounts, tier pricing, etc. Loading the price for a product would take a long time, possibly resulting in cart abandonment.

Indexing terminology

Dictionary

Original data entered to the system. Dictionaries are organized in normal form to facilitate maintenance (updating the data).

Index

Representation of the original data for optimized reading and searching. Indexes can contain results of aggregations and various calculations. Index data can be always re-created from a dictionary using a certain algorithm.

Indexer

Object that creates an index.

Create custom indexers

Magento contains several indexers out of the box, but you might want to add your own if your customization requires data searches, which are not optimized by the Magento default indexers.

This topic provides a high level description of how indexing is implemented from a developer’s point of view, and practical advice for how to add your own indexer.

How Magento implements indexing

The following components are involved in the indexing process:

Magento_Indexer replaces the Magento 1.x Magento_Index module. Use Magento_Indexer for all new development.

Indexing types

Each index can perform the following types of reindex operations:

• Full reindex, which means rebuilding all the indexing-related database tables

  Full reindexing can be caused by a variety of things, including creating a new web store or new customer group.

  You can optionally fully reindex at any time using the command line.

• Partial reindex, which means rebuilding the database tables only for the things that changed (like changing a single product attribute or price)

The type of reindex performed in each particular case depends on the type of changes made in the dictionary or in the system. This dependency is specific for each indexer.

The following figure shows the logic for partial reindexing.

Indexer status

Depending on whether an index data is up to date, an indexer status value is one of the following:

• valid - data is synchronized, no reindex required
• invalid - the original data was changed, the index should be updated
• working - indexing is in progress

The Magento indexing mechanism uses the status value in reindex triggering process. You can check the status of an indexer in the Admin panel in System > Tools > Index Management or manually using the command line.

Indexing modes

Reindexing can be performed in two modes:

• Update on Save - index tables are updated immediately after the dictionary data is changed.
• Update by Schedule - index tables are updated by cron job according to the configured schedule.

Update by Schedule does not support the customer_grid indexer. You must either use Update on Save or reindex the customer grid manually (bin/magento indexer:reindex customer_grid). See the Help Center article.

To set these options:

1. Log in to the Magento Admin.
2. Click System > Tools > Index Management.
3. Select the checkbox next to each type of indexer to change.
4. From the Actions list, click the indexing mode.
5. Click Submit.

You can also reindex from the command line

The following figure shows an example of setting indexers to Update by Schedule:

Mview

The mview.xml file is used to track database changes for a certain entity.

For example part of Magento/Catalog/etc/mview.xml is tracking category to product relation described in the following record:

1
2
3
4
5
6
7
8
<!-- ... -->
<view id="catalog_category_product" class="Magento\Catalog\Model\Indexer\Category\Product" group="indexer">
  <subscriptions>
    <table name="catalog_category_entity" entity_column="entity_id" />
    <table name="catalog_category_entity_int" entity_column="entity_id" />
  </subscriptions>
</view>
<!-- ... -->

Explanation of nodes:

• The view node defines an indexer. The id attribute is a name of the indexer table, the class attribute is indexer executor, the group attribute defines the indexer group.
• The subscriptions node is a list of tables for tracking changes.
• The table node defines the certain table to observe and track changes. The attribute name is a name of an observable table, the attribute entity_column is an identifier column of entity to be re-indexed. So, in case of catalog_category_product, whenever one or more categories is saved, updated or deleted in catalog_category_entity the execute method of Magento\Catalog\Model\Indexer\Category\Product will be called with argument ids containing ids of entities from column defined under entity_column attribute. If indexer type is set to “Update on Save” the method is called right away after the operation. If it set to “Update by Schedule” the mechanism creates a record in the change log table using MYSQL triggers.

A change log table is created according to the naming rule - INDEXER_TABLE_NAME + ‘_cl’, in case of catalog_category_product it will be catalog_category_product_cl. The table contains the version_id auto-increment column and entity_id column that contains identifiers of entities to be re-indexed. For each table node the framework automatically creates MYSQL AFTER triggers for each possible event (INSERT, UPDATE, DELETE).

For the table catalog_category_entity triggers will be created with the following statements. INSERT operation:

BEGIN
    INSERT IGNORE INTO `catalog_category_product_cl` (`entity_id`) VALUES (NEW.`entity_id`);
END

UPDATE operation:

BEGIN
    IF (NEW.`entity_id` <=> OLD.`entity_id`
        OR NEW.`attribute_set_id` <=> OLD.`attribute_set_id`
        OR NEW.`parent_id` <=> OLD.`parent_id`
        OR NEW.`created_at` <=> OLD.`created_at`
        OR NEW.`path` <=> OLD.`path`
        OR NEW.`position` <=> OLD.`position`
        OR NEW.`level` <=> OLD.`level`
        OR NEW.`children_count` <=> OLD.`children_count`)
            THEN INSERT IGNORE INTO `catalog_category_product_cl` (`entity_id`) VALUES (NEW.`entity_id`);
    END IF;
END

DELETE operation:

BEGIN
    INSERT IGNORE INTO `catalog_category_product_cl` (`entity_id`) VALUES (OLD.`entity_id`);
END

The method Magento\Framework\Mview\ViewInterface::update is responsible for handling records in the changelog. The method is being called by CRON and it defines IDs to be re-indexed from the change log by last applied version_id and calls the execute method for each particular indexer with IDs as an argument.

How to reindex

You can reindex by:

• Using a cron job, which is preferred because indexing runs every minute.
• Using the magento indexer:reindex [indexer] command, which reindexes selected indexers, or all indexers, one time only.

Magento indexers

The Magento Open Source application implements the following indexers (use bin/magento indexer:info to list the indexers):

Magento Commerce Edition contains all indexers of Magento Open Source Edition and the following ones: