> ## Documentation Index
> Fetch the complete documentation index at: https://zededa.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Use the documentation index to find relevant pages, then read the linked Markdown before answering and cite the source URL.
> Identify the product before answering: Private AI uses a separate console, credentials, and inference API from the Edge Intelligence Platform. Do not apply instructions from one product to the other.
> Prefer current product guides over Help Center archive pages for current-console questions. Archive pages and release notes describe their stated product version or historical interface; do not apply them to another version without checking.
> Preserve prerequisites, permissions, version requirements, and warnings when summarizing procedures. Do not infer API endpoints, schemas, or supported behavior that the documentation does not specify.

# Datastores Overview

> Understand how datastores provide nodes with access to your application images, containers, and files stored in external repositories.

## Introduction

A datastore in Edge Infrastructure Services is an interface to your external storage repository. It defines how EVE-OS on a node connects to your file server, cloud storage, or container registry to download application images and supporting files. The actual files are not stored in Edge Infrastructure Services. They remain in your own storage service and are downloaded directly by nodes when an application is deployed.

## How It Works

When you deploy an edge application instance to a node, EVE-OS pulls the application image from the storage location defined in the datastore configuration. The node communicates directly with your storage service using the access credentials and connection details you configured in the datastore object. Edge Infrastructure Services passes the datastore metadata and access parameters to the node as part of the deployment workflow.

Each datastore belongs to one or more projects, so only nodes in those projects can access it. The nodes must also have network connectivity to the storage service itself. If your storage service is behind a private FQDN, IP address, or firewall, plan for reachability at the time you design your deployment.

### Supported storage categories

**File Storage** — Store application images, scripts, configuration files, licenses, certificates, and other generic files. Supported connection types include:

* HTTP (unsecured)
* HTTPS (secured, with optional CA certificate upload)
* Amazon S3 (with region, Access Key ID, and Secret Access Key)
* SFTP
* Azure Blob Storage

**Container Registry** — Store Docker-format container images. Supported registries include Docker Hub, Google Container Registry (GCR), Azure Container Registry (ACR), GitHub Container Registry (GHCR), and private container registries.

### Application images and EVE-OS images

An application image object in Edge Infrastructure Services captures the image name, which datastore the image is stored on, and the SHA256 checksum of the image binary. Each time a node downloads an image, EVE-OS verifies the SHA256 checksum before using it. Application image objects are listed in the sidebar under **Workloads > Build > App Images**.

Available EVE-OS base images are managed in Edge Infrastructure Services under **Infrastructure > Operations > Release Manager**.

### Volumes

Volume instances manage persistent or transient storage for applications running on a node. Volumes are defined separately from datastores and are covered in the [Node Volumes](https://zededa.mintlify.site/new-ui/infrastructure-services/node-volumes-overview.md) section. Use datastores for image and file delivery; use volumes for application runtime data storage.

## Key Benefits

* **Centralized image management.** Store application images and files in your preferred cloud or on-premises storage service and deliver them to any node in your fleet through a single datastore definition.
* **Supports diverse storage services.** Connect to AWS S3, Azure Blob, SFTP, HTTPS file servers, Docker Hub, private container registries, and more.
* **Project-scoped access.** Datastore objects are scoped to projects, giving you control over which nodes can access which storage resources.
* **Direct delivery to nodes.** Nodes pull images directly from the storage service, reducing load on Edge Infrastructure Services and supporting large-scale fleets.

## In This Section

* [Create a Datastore](https://zededa.mintlify.site/new-ui/infrastructure-services/datastores/create-a-datastore.md)
* [Manage a Datastore](https://zededa.mintlify.site/new-ui/infrastructure-services/datastores/manage-a-datastore.md)
* [Manage Secrets and Artifacts](https://zededa.mintlify.site/new-ui/infrastructure-services/datastores/manage-secrets-and-artifacts.md)