Choose a service

Applies to VAST Cluster 5.5.

Find what you want to do under the service that does it, then start with that service's overview and quickstart.

VAST DataBase

I want to …	Start with
Store tables and query them with SQL or from Python, with transactions and row-level updates and deletes	Overview, quickstart
Run several engines (Python, Trino, Spark) on one copy of the same tables	Overview, quickstart

VAST Vector Search

Overview · Quickstart

I want to …	Start with
Store embeddings and run nearest-neighbour search, with SQL filters	Overview, quickstart
Retrieve document chunks for RAG, returning only chunks the user may read	RAG retrieval, quickstart

VAST Event Broker

Overview · Quickstart

I want to …	Start with
Produce and consume streaming events with Kafka clients	Overview, quickstart
Query a stream's events with SQL, without a connector job that copies them into tables	Query topics as tables, quickstart
Parse JSON events into typed columns of a table as they arrive	Blob expansion
Get a Kafka event each time an object in an S3 bucket is created, tagged or deleted, for your own consumer	S3 bucket events

VAST DataEngine

Overview · Quickstart

I want to …	Start with
Run Python code each time an object or file is created or deleted, or on a schedule	Overview, quickstart
Compute embeddings, or write rows to a table, as files and objects arrive	VAST DataBase and Vector Search from a function

VAST DataStore

Overview · Quickstart

I want to …	Start with
Read and write objects over S3 with boto3, the AWS CLI, s3fs or PyArrow	Overview, quickstart
Mount files over NFS for an application, or share the same data over S3 and NFS	NFS for applications, multiprotocol access

VAST Catalog

Overview · Quickstart

I want to …	Start with
Find files and objects by name, size, age, owner or S3 tag without walking directories	Overview, quickstart
See what changed in a namespace between two points in time	Snapshots and change detection
Find out who read, wrote or deleted a path, and whether it succeeded (audit log)	Query the audit log

Across services

I want to …	Start with
Create views, policies, keys, tables or topics from CI	Automation tools compared

Some things to know before you choose:

Search runs through the VAST Query Engine. The Trino and Spark connectors don't support vector index-based search (VAST Vector Search overview).
DataEngine is for per-event or per-batch work, not for SQL over large tables. For that, use the VAST Query Engine, Trino or Spark (VAST DataEngine overview).
You don't need DataEngine just to get S3 bucket notifications into a Kafka topic for your own consumers (VAST DataEngine overview).
The Catalog lags the file system. It's as fresh as its last indexed snapshot, so don't use it for anything that must see a file the moment it's written (VAST Catalog overview).

Combining services

Many applications use more than one service: for example, objects landing in DataStore trigger a DataEngine function that writes to a VAST DataBase table. End-to-end patterns outlines the common combinations and links each step to the page that covers it.