Glossary of termsedit


Analysis is the process of converting full text to terms. Depending on which analyzer is used, these phrases: FOO BAR, Foo-Bar, foo,bar will probably all result in the terms foo and bar. These terms are what is actually stored in the index.

A full text query (not a term query) for FoO:bAR will also be analyzed to the terms foo,bar and will thus match the terms stored in the index.

It is this process of analysis (both at index time and at search time) that allows Elasticsearch to perform full text queries.

Also see text and term.

API key
A unique identifier that you can use for authentication when submitting Elasticsearch requests. When TLS is enabled, all requests must be authenticated using either basic authentication (user name and password) or an API key.
auto-follow pattern
An index pattern that automatically configures new indices as follower indices for cross-cluster replication. For more information, see Managing auto follow patterns.
One or more nodes that share the same cluster name. Each cluster has a single master node, which is chosen automatically by the cluster and can be replaced if it fails.
cold phase
The third possible phase in the index lifecycle. In the cold phase, an index is no longer updated and seldom queried. The information still needs to be searchable, but it’s okay if those queries are slower.
cold tier
A data tier that contains nodes that hold time series data that is accessed occasionally and not normally updated.
component template
A building block for constructing index templates that specifies index mappings, settings, and aliases.
content tier
A data tier that contains nodes that handle the indexing and query load for content such as a product catalog.
cross-cluster replication (CCR)
A feature that enables you to replicate indices in remote clusters to your local cluster. For more information, see Cross-cluster replication.
cross-cluster search (CCS)
The cross-cluster search feature enables any node to act as a federated client across multiple clusters. See Search across clusters.
data stream

A named resource used to ingest, search, and manage time series data in Elasticsearch. A data stream’s data is stored across multiple hidden, auto-generated indices. You can automate management of these indices to more efficiently store large data volumes.

See Data streams.

data tier
A collection of nodes with the same data role that typically share the same hardware profile. See content tier, hot tier, warm tier, cold tier.
delete phase
The last possible phase in the index lifecycle. In the delete phase, an index is no longer needed and can safely be deleted.

A document is a JSON document which is stored in Elasticsearch. It is like a row in a table in a relational database. Each document is stored in an index and has a type and an id.

A document is a JSON object (also known in other languages as a hash / hashmap / associative array) which contains zero or more fields, or key-value pairs.

The original JSON document that is indexed will be stored in the _source field, which is returned by default when getting or searching for a document.


A document contains a list of fields, or key-value pairs. The value can be a simple (scalar) value (eg a string, integer, date), or a nested structure like an array or an object. A field is similar to a column in a table in a relational database.

The mapping for each field has a field type (not to be confused with document type) which indicates the type of data that can be stored in that field, eg integer, string, object. The mapping also allows you to define (amongst other things) how the value for a field should be analyzed.

A filter is a non-scoring query, meaning that it does not score documents. It is only concerned about answering the question - "Does this document match?". The answer is always a simple, binary yes or no. This kind of query is said to be made in a filter context, hence it is called a filter. Filters are simple checks for set inclusion or exclusion. In most cases, the goal of filtering is to reduce the number of documents that have to be examined.
Peform a Lucene commit to write index updates in the transaction log (translog) to disk. Because a Lucene commit is a relatively expensive operation, Elasticsearch records index and delete operations in the translog and automatically flushes changes to disk in batches. To recover from a crash, operations that have been acknowledged but not yet committed can be replayed from the translog. Before upgrading, you can explicitly call the Flush API to ensure that all changes are committed to disk.
follower index
The target index for cross-cluster replication. A follower index exists in a local cluster and replicates a leader index.
force merge
Manually trigger a merge to reduce the number of segments in each shard of an index and free up the space used by deleted documents. You should not force merge indices that are actively being written to. Merging is normally performed automatically, but you can use force merge after rollover to reduce the shards in the old index to a single segment. See the force merge API.
Make an index read-only and minimize its memory footprint. Frozen indices can be searched without incurring the overhead of of re-opening a closed index, but searches are throttled and might be slower. You can freeze indices to reduce the overhead of keeping older indices searchable before you are ready to archive or delete them. See the freeze API.
frozen index
An index reduced to a low overhead state that still enables occasional searches. Frozen indices use a memory-efficient shard implementation and throttle searches to conserve resources. Searching a frozen index is lower overhead than re-opening a closed index to enable searching.
hidden index
An index that is excluded by default when you access indices using a wildcard expression. You can specify the expand_wildcards parameter to include hidden indices. Note that hidden indices are included if the wildcard expression starts with a dot, for example .watcher-history*.
hot phase
The first possible phase in the index lifecycle. In the hot phase, an index is actively updated and queried.
hot tier
A data tier that contains nodes that handle the indexing load for time series data such as logs or metrics and hold your most recent, most-frequently-accessed data.
The ID of a document identifies a document. The index/id of a document must be unique. If no ID is provided, then it will be auto-generated. (also see routing)

An optimized collection of JSON documents. Each document is a collection of fields, the key-value pairs that contain your data.

An index is a logical namespace which maps to one or more primary shards and can have zero or more replica shards.

index alias

An index alias is a secondary name used to refer to one or more existing indices.

Most Elasticsearch APIs accept an index alias in place of an index name.

See Add index alias.

index lifecycle
The four phases an index can transition through: hot, warm, cold, and delete. For more information, see Index lifecycle.
index lifecycle policy
Specifies how an index moves between phases in the index lifecycle and what actions to perform during each phase.
index pattern
A string that can contain the * wildcard to match multiple index names. In most cases, the index parameter in an Elasticsearch request can be the name of a specific index, a list of index names, or an index pattern. For example, if you have the indices datastream-000001, datastream-000002, and datastream-000003, to search across all three you could use the datastream-* index pattern.
index template

Defines settings and mappings to apply to new indexes that match a simple naming pattern, such as logs-*. An index template can also attach a lifecycle policy to the new index. Index templates are used to automatically configure indices created during rollover.

leader index
The source index for cross-cluster replication. A leader index exists on a remote cluster and is replicated to follower indices.
local cluster
The cluster that pulls data from a remote cluster in cross-cluster search or cross-cluster replication.

A mapping is like a schema definition in a relational database. Each index has a mapping, which defines a type, plus a number of index-wide settings.

A mapping can either be defined explicitly, or it will be generated automatically when a document is indexed.


A node is a running instance of Elasticsearch which belongs to a cluster. Multiple nodes can be started on a single server for testing purposes, but usually you should have one node per server.

At startup, a node will use unicast to discover an existing cluster with the same cluster name and will try to join that cluster.

primary shard

Each document is stored in a single primary shard. When you index a document, it is indexed first on the primary shard, then on all replicas of the primary shard.

By default, an index has one primary shard. You can specify more primary shards to scale the number of documents that your index can handle.

You cannot change the number of primary shards in an index, once the index is created. However, an index can be split into a new index using the split API.

See also routing


A request for information from Elasticsearch. You can think of a query as a question, written in a way Elasticsearch understands. A search consists of one or more queries combined.

There are two types of queries: scoring queries and filters. For more information about query types, see Query and filter context.


Shard recovery is the process of syncing a replica shard from a primary shard. Upon completion, the replica shard is available for search.

Recovery automatically occurs during the following processes:


Copies documents from a source to a destination. The source and destination can be any pre-existing index, index alias, or data stream.

You can reindex all documents from a source or select a subset of documents to copy. You can also reindex to a destination in a remote cluster.

A reindex is often performed to update mappings, change static index settings, or upgrade Elasticsearch between incompatible versions.

remote cluster
A separate cluster, often in a different data center or locale, that contains indices that can be replicated or searched by the local cluster. The connection to a remote cluster is unidirectional.
replica shard

Each primary shard can have zero or more replicas. A replica is a copy of the primary shard, and has two purposes:

  1. increase failover: a replica shard can be promoted to a primary shard if the primary fails
  2. increase performance: get and search requests can be handled by primary or replica shards.

    By default, each primary shard has one replica, but the number of replicas can be changed dynamically on an existing index. A replica shard will never be started on the same node as its primary shard.


Creates a new index for a rollover target when the existing index reaches a certain size, number of docs, or age. A rollover target can be either an index alias or a data stream.

For example, if you’re indexing log data, you might use rollover to create daily or weekly indices. See the rollover index API.

Summarize high-granularity data into a more compressed format to maintain access to historical data in a cost-effective way.
rollup index
A special type of index for storing historical data at reduced granularity. Documents are summarized and indexed into a rollup index by a rollup job.
rollup job
A background task that runs continuously to summarize documents in an index and index the summaries into a separate rollup index. The job configuration controls what information is rolled up and how often.

When you index a document, it is stored on a single primary shard. That shard is chosen by hashing the routing value. By default, the routing value is derived from the ID of the document or, if the document has a specified parent document, from the ID of the parent document (to ensure that child and parent documents are stored on the same shard).

This value can be overridden by specifying a routing value at index time, or a routing field in the mapping.

searchable snapshot
A snapshot of an index that has been mounted as a searchable snapshot index and can be searched as if it were a regular index.
searchable snapshot index
An index whose data is stored in a snapshot that resides in a separate snapshot repository such as AWS S3. Searchable snapshot indices do not need replica shards for resilience, since their data is reliably stored outside the cluster.

A shard is a single Lucene instance. It is a low-level “worker” unit which is managed automatically by Elasticsearch. An index is a logical namespace which points to primary and replica shards.

+ Other than defining the number of primary and replica shards that an index should have, you never need to refer to shards directly. Instead, your code should deal only with an index.

+ Elasticsearch distributes shards amongst all nodes in the cluster, and can move shards automatically from one node to another in the case of node failure, or the addition of new nodes.

Reduce the number of primary shards in an index. You can shrink an index to reduce its overhead when the request volume drops. For example, you might opt to shrink an index once it is no longer the write index. See the shrink index API.
Captures the state of the whole cluster or of particular indices or data streams at a particular point in time. Snapshots provide a back up of a running cluster, ensuring you can restore your data in the event of a failure. You can also mount indices or datastreams from snapshots as read-only searchable snapshots.
snapshot lifecycle policy
Specifies how frequently to perform automatic backups of a cluster and how long to retain the resulting snapshots.
snapshot repository
Specifies where snapshots are to be stored. Snapshots can be written to a shared filesystem or to a remote repository.
source field
By default, the JSON document that you index will be stored in the _source field and will be returned by all get and search requests. This allows you access to the original object directly from search results, rather than requiring a second step to retrieve the object from an ID.
system index
An index that contains configuration information or other data used internally by the system, such as the .security index. The name of a system index is always prefixed with a dot. You should not directly access or modify system indices.

A term is an exact value that is indexed in Elasticsearch. The terms foo, Foo, FOO are NOT equivalent. Terms (i.e. exact values) can be searched for using term queries.

See also text and analysis.


Text (or full text) is ordinary unstructured text, such as this paragraph. By default, text will be analyzed into terms, which is what is actually stored in the index.

Text fields need to be analyzed at index time in order to be searchable as full text, and keywords in full text queries must be analyzed at search time to produce (and search for) the same terms that were generated at index time.

See also term and analysis.

A type used to represent the type of document, e.g. an email, a user, or a tweet. Types are deprecated and are in the process of being removed. See Removal of mapping types.
warm phase
The second possible phase in the index lifecycle. In the warm phase, an index is generally optimized for search and no longer updated.
warm tier
A data tier that contains nodes that hold time series data that is accessed less frequently and rarely needs to be updated.