Collection

On Zilliz Cloud, a collection is a table used to hold vector embeddings and their metadata. In this chapter, you will learn the definition and related concepts of the collection with relevant demonstrations.

Overview [READ MORE]

On Zilliz Cloud, you can create multiple collections to manage your data, and insert your data as entities into the collections. Collections and entities are similar to tables and records in relational databases. This page helps you to learn about the collection and related concepts.

Create [READ MORE]

You can create a collection by defining its schema, index parameters, metric type, and whether to load it upon creation. This page introduces how to create a collection from scratch.

View [READ MORE]

You can obtain the name list of all the collections in the currently connected database, and check the details of a specific collection.

Modify [READ MORE]

You can rename a collection or change its settings. This page focuses on how to modify a collection.

TTL [READ MORE]

Zilliz Cloud can automatically expire entities through a Time-to-Live (TTL) policy. Expired entities stop appearing in query and search results immediately, and are physically removed from storage on the next compaction cycle — typically within 24 hours.

Load & Release [READ MORE]

Loading a collection is a prerequisite for conducting similarity searches and queries within it. This page focuses on the procedures for loading and releasing a collection.

Partitions [READ MORE]

A partition is a subset of a collection. Each partition shares the same data structure with its parent collection but contains only a subset of the data in the collection. This page helps you understand how to manage partitions.

Aliases [READ MORE]

In Zilliz Cloud, an alias is a secondary, mutable name for a collection. Using aliases provides a layer of abstraction that allows you to dynamically switch between collections without modifying your application code. This is particularly useful in production environments for seamless data updates, A/B testing, and other operational tasks.

Drop [READ MORE]

You can drop a collection if it is no longer needed.

On Console [READ MORE]

A collection is a two-dimensional table used to store vector embeddings and metadata. All entities in a collection share the same schema. You can create multiple collections for data management or multi-tenancy purposes.

Large TopK [READ MORE]

A Zilliz Cloud collection allows you to retrieve up to 16,384 entities in a search or query result. To retrieve more entities beyond the topK limit, you can set the query mode to allow Zilliz Cloud to include millions of entities in a single search or query result, instead of using complex and time-consuming iterators.