---
title: Managing Collections
description: Managing Collections with the Analytics Workbench.
pubDate: 2026-08-17T09:53:44.266Z
antora:
  editUrl: https://github.com/couchbase/docs-analytics/edit/release/8.0/modules/analytics/pages/manage-datasets.adoc
  xref: xref:server:analytics:manage-datasets.adoc[]
---

[Consult the llms.txt file for a full list of contents](/llms.txt)
[View original HTML](/server/current/analytics/manage-datasets.html)

# Managing Collections

The Analytics Workbench enables you to create, edit, and drop collections. The Analytics Workbench also enables you to connect or disconnect local links and remote links.

Collections are displayed in the insights sidebar of the Analytics Workbench. Each collection is listed below the link on which it was created.

![The insights sidebar with collections displayed](_images/workbench-insights-datasets.png) 

Refer to [Analytics Collections](5%5Fddl.md#Datasets) for further details on collections.

## [](#mapping-collections-from-the-data-service)Mapping Collections from the Data Service

Mapping collections from the Data service provides a quick way to make collections in the Data service available in Analytics. The mapping process creates an Analytics scope that corresponds to each specified Data service scope, and within each Analytics scope, an Analytics collection that corresponds to each specified Data service collection. Each mapped Analytics collection uses the equivalent Data service collection as its data source.

To map collections from the Data service:

1. In the insights sidebar, click the **Map From Data Service** button.  
The **Map Collections From Data Service** dialog is displayed, showing all the buckets available in the Data service.  
![The Map Collections From Data Service dialog](_images/dataset-add-local.png)
2. Locate and select each collection in the Data service that you want to map:

  1. Click the rightward-pointing arrowhead  before the name of the required bucket, to display the scopes within that bucket.
  2. Click the rightward-pointing arrowhead  before the name of the required scope, to display the collections within that scope.
  3. Check the box next to each of the collections that you want to map.
3. Choose **Save Mapping** to map the selected collections, or **Cancel** to cancel.

You can also map collections from the Data service using the ALTER COLLECTION statement. For more information about mapping collections from the Data Service, refer to [Alter Statements](5%5Fddl.md#Alter%5Fstatements).

## [](#creating-a-collection-on-a-local-link)Creating a Collection on a Local Link

Although mapping collections from the Data service is convenient, you may sometimes need greater control over the creation of an Analytics collection on a local link; for example, if you need to create an Analytics collection which has a different name or different scope to its data source. In this case, you can create a single Analytics collection on a local link.

To create an Analytics collection on a local link:

1. In the insights sidebar, click **\+ collection** next to the local link where you want to create the collection.  
The **Add Analytics Collection** dialog is displayed.  
![The Add Analytics Collection dialog](_images/dataset-add-remote.png)
2. In the **Collection Name** box, enter a name for the Analytics collection.
3. Use the **Source** drop-down lists to select a bucket, scope, and collection to specify the data source for this Analytics collection.
4. If necessary, in the **Where** box, enter an expression to filter the documents in the collection. You must not include the WHERE keyword, and the expression must be deterministic. For example, `activity = "eat"`.
5. Choose **Save** to create the link, or **Cancel** to cancel.

You can also create a collection on a local link using the CREATE ANALYTICS COLLECTION statement. For more information about creating collections, refer to [Create Statements](5%5Fddl.md#Creating%5Fand%5Fdropping).

## [](#creating-a-collection-on-a-remote-link)Creating a Collection on a Remote Link

To create an Analytics collection on a remote link:

1. In the insights sidebar, click **\+ collection** next to the remote link where you want to create the collection.  
The **Add Analytics Collection** dialog is displayed.  
![The Add Analytics Collection dialog](_images/dataset-add-remote.png)
2. In the **Collection Name** box, enter a name for the collection.
3. Use the **Source** drop-down lists to select a bucket, scope, and collection on the remote Couchbase cluster which specify the data source for this Analytics collection.
4. If necessary, in the **Where** box, enter an expression to filter the documents in the collection. You must not include the WHERE keyword, and the expression must be deterministic. For example, `activity = "eat"`.
5. Choose **Save** to create the link, or **Cancel** to cancel.

You can also create a collection on a remote link using the CREATE ANALYTICS COLLECTION statement. For more information about creating collections, refer to [Create Statements](5%5Fddl.md#Creating%5Fand%5Fdropping).

## [](#creating-a-collection-on-an-external-link)Creating a Collection on an External Link

To create an Analytics collection on an external link:

1. In the insights sidebar, click **\+ collection** next to the external link where you want to create the collection.  
The **Add Analytics Collection** dialog is displayed.  
![The Add Analytics Collection dialog with options for an external link](_images/dataset-add-external.png)
2. In the **Collection Name** box, enter a name for the collection.
3. In the **Bucket Name** box, enter the name of the external data source for this collection, such as an Amazon S3 bucket.
4. If necessary, in the **Path** box, enter the path to the location of the data relative to the external data source. For example: `json-data/reviews`.
5. Open the **File Format** drop-down list and select the format of the external data:

  * **JSON** — Data is in [JSON Lines](http://jsonlines.org) format.
  * **CSV** — Data is in [comma-separated values](https://tools.ietf.org/html/rfc4180) format.
  * **TSV** — Data is in [tab-separated values](https://www.iana.org/assignments/media-types/text/tab-separated-values) format.
  * **Parquet** — Data is in [Apache Parquet](https://parquet.apache.org) format.  
Note that is currently only supported on the Amazon S3 service.
6. If you specified **CSV** or **TSV** file format, in the **Type definition** box, enter a comma-separated list of field definitions to guide the transformation of each CSV or TSV record into a JSON object. Each field definition should consist of:

  * A name to be assigned to the field.
  * The data type of the field. This may be any of the [primitive data types](10%5Fdata%5Ftype.md#PrimitiveTypes), where `INT` is an alias for `BIGINT`. If this field does not contain a value of this data type, the record is ignored.
  * Optionally, the `NOT UNKNOWN` flag. When this flag is present, if this field is `missing` or `null`, the record is ignored.  
For example: `id INT NOT UNKNOWN, datetime STRING NOT UNKNOWN, amount DOUBLE NOT UNKNOWN, details STRING`.
7. If you specified **CSV** or **TSV** file format, check the **Header** check box if the first row of the file should be skipped.
8. If you specified **CSV** or **TSV** file format, in the **Null Value** box, enter the string used in the external collection to represent a `null` value. You may escape special characters using a backslash `\`.
9. In the **Include** box, enter the names of the files to be included from the external collection. The file name may include subdirectories of the location specified by the **Path** option. Any files that do not match the specification are excluded. This option may not be used if the **Exclude** option is specified.  
The file names may contain the following wildcard characters, in common with the **Exclude** option:  
`*` — Matches anything  
`?` — Matches any single character  
`[` _sequence_ `]` — Matches any characters in _sequence_  
`[!` _sequence_ `]` — Matches any characters not in _sequence_  
For example: `["*2018*.json", "*2019*.json"]`.
10. In the **Exclude** box, enter the names of the files to be excluded from the external collection. The file name may include subdirectories of the location specified by the **Path** option. Any files that do not match the specification are included. This option may not be used if the **Include** option is specified.  
The file names may contain the same wildcard characters as the **Include** option. For example: `"*.?sv"`.
11. Choose **Save** to create the link, or **Cancel** to cancel.

Note that for an external collection, data in JSON, CSV, or TSV format may be stored in compressed GZIP files, with the extension `.gz` or `.gzip`.

You can also create a collection on an external link using the CREATE EXTERNAL ANALYTICS COLLECTION statement. For more information about creating collections, refer to [Create Statements](5%5Fddl.md#Creating%5Fand%5Fdropping).

## [](#connecting-collections)Connecting Collections

When you have created one or more collections on a local link or a remote link, you must connect the link, so that collections on the link can shadow data from their data sources.

To connect a link, click the disconnected link icon  next to the link name in the insights sidebar.

When the link is connected, a connected link icon is displayed next to the link name, and any collections listed below that link are shown in black.

![Connected link in the insights sidebar](_images/workbench-connected-link.png) 

You can also connect a link using the CONNECT statement. For more information about connecting links, refer to [Connect Statements](5%5Fddl.md#Connecting%5Fand%5Fdisconnecting).

## [](#disconnecting-collections)Disconnecting Collections

Disconnecting a local link or a remote link disconnects all collections on that link from their data sources.

To disconnect a link, click the connected link icon  next to the link name in the insights sidebar.

When the link is disconnected, a disconnected link icon is displayed next to the link name, and any collections listed below that link are shown in orange.

![Disconnected link in the insights sidebar](_images/workbench-disconnected-link.png) 

You can also disconnect a link using the DISCONNECT statement. For more information on disconnecting links, refer to [Disconnect Statements](5%5Fddl.md#Disconnect%5Fstatements).

## [](#viewing-collection-summaries)Viewing Collection Summaries

You can view a summary of a collection in the insights sidebar.

To display a summary of a collection, click the collection name in the insights sidebar, or click the rightward-pointing arrowhead  before the collection name.

The collection summary appears below the collection name, showing the name of the keyspace which is the data source for this collection, and (if applicable) the expression which filters the documents in the keyspace.

If you have created any indexes on this collection, the index names and definitions are shown below the collection summary.

![Collection summary in the insights sidebar](_images/workbench-dataset-summary.png) 

To hide the summary of a collection, click the collection name in the insights sidebar again, or click the downward-pointing arrowhead  before the collection name.

When data ingestion is in progress, an indicator below each collection shows the percentage of mutations that remain to be synchronized to that collection. If the indicator is not displayed, then the collection is up-to-date.

![Collection showing ingestion progress](_images/workbench-dataset-ingestion.png) 

## [](#deleting-a-collection)Deleting a Collection

To delete an Analytics collection:

1. If necessary, in the insights sidebar, click the name of the collection to display the collection summary.
2. Click **drop analytics collection** below the collection name.  
A message box appears, asking if you are sure that you want to delete the collection.
3. Choose **Continue** to delete the collection, or **Cancel** to cancel.

You can also delete a collection using the DROP ANALYTICS COLLECTION statement. For more information on deleting a collection, refer to [Drop Statements](5%5Fddl.md#Drop%5Fstatements).

## [](#deleting-a-scope)Deleting a Scope

You can delete an Analytics scope from the insights sidebar, as long as the scope contains no remote links or external links, and contains no collections on the Local link. You cannot delete the `Default` Analytics scope.

To delete an Analytics scope:

1. In the insights sidebar, click the trash can icon  next to the scope name.  
A message box appears, asking if you are sure that you want to delete the scope.
2. Choose **Continue** to delete the scope, or **Cancel** to cancel.