> For the complete documentation index, see [llms.txt](https://docs.silk.us/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.silk.us/deploy/plan-your-deployment/plan-a-silk-deployment.md).

# Plan a Silk Deployment

Understand the components and planning decisions involved in deploying Silk, then choose the deployment path for your cloud provider.

A Silk deployment brings together cloud infrastructure, Silk Flex, a Silk cluster, and one or more DataPods to serve your application data. The overall path is similar across AWS, Azure and Google Cloud: plan your environment, prepare the required cloud resources, deploy Flex, create a cluster, and provision DataPods.

However, the specific resources, permissions, networking configuration, and installation steps differ by cloud provider. Use this article to understand the core deployment decisions, then follow the learning path for your environment.

## Understand the main deployment components

{% columns %}
{% column %}
**Silk Flex** is the orchestration and management layer. It is deployed in your cloud environment and manages the creation and operation of Silk clusters and DataPods.
{% endcolumn %}

{% column %}
A **Silk cluster** enables the configuration and resources within which DataPods operate. Its configuration depends on deployment inputs like networking and planned size.
{% endcolumn %}

{% column %}
**DataPods** provide the storage resources that serve application data. After creating a cluster, provision DataPods and prepare storage/host connectivity for your workloads.
{% endcolumn %}
{% endcolumns %}

Your deployment also depends on **cloud infrastructure and application connectivity**: the cloud provider's compute resources, networking, identities and permissions, and a network path between application hosts and DataPods.

## Make the key planning decisions

Before configuring cloud resources, work with Silk's Cloud Infrastructure team on the pre-installation site survey, which establishes the deployment inputs and confirms that the proposed environment supports your intended configuration.

<table><thead><tr><th width="280">Planning area</th><th>What to establish</th></tr></thead><tbody><tr><td><strong>Location and cloud environment</strong></td><td>The target account, subscription, or project, and the region or zone where applicable.</td></tr><tr><td><strong>Cluster and DataPods</strong></td><td>The planned cluster size, initial DataPod configuration, workload needs, and room for growth.</td></tr><tr><td><strong>Resource availability</strong></td><td>Availability of the required compute resources, images, and cloud quotas.</td></tr><tr><td><strong>Networking and connectivity</strong></td><td>Network and address-space design, required traffic paths, and connectivity from application hosts.</td></tr><tr><td><strong>Access and permissions</strong></td><td>The identities, roles, and permissions needed for deployment and ongoing operation.</td></tr></tbody></table>

The site survey should be finalized before requesting the cluster file used for cluster creation. See the [Plan DataPod Capacity](/deploy/plan-your-deployment/plan-datapod-capacity.md) and [Cluster and DataPod Sizing](/deploy/plan-your-deployment/cluster-and-datapod-sizing.md) documentation for detailed sizing guidance.

## Compare deployments by cloud provider

Silk supports deployments on AWS, Azure, and Google Cloud. The following table summarizes the differences that affect how you plan and prepare each environment. For architecture details and deployment steps, see the overview for your cloud provider.

<table data-search="false"><thead><tr><th width="180">Deployment area</th><th>AWS</th><th>Azure</th><th>Google Cloud</th></tr></thead><tbody><tr><td><strong>Deployment scope</strong></td><td>AWS account and region</td><td>Subscription, region, and a dedicated, empty Flex resource group</td><td>Google Cloud project, region, and zone</td></tr><tr><td><strong>Flex installation</strong></td><td>AWS Marketplace using AWS CloudFormation</td><td>Azure Marketplace</td><td>Google Cloud Marketplace using Infrastructure Manager</td></tr><tr><td><strong>Deployment models</strong></td><td>Single deployment model</td><td>Flex-managed mode (SMI) or customer-managed mode (UMI)</td><td>Single deployment model</td></tr><tr><td><strong>Network layout</strong></td><td>Six dedicated subnets in one existing VPC</td><td>Six dedicated subnets in one VNet, which can be dedicated or shared with application hosts</td><td>Six dedicated VPC networks, each with its own subnet</td></tr><tr><td><strong>Network creation</strong></td><td>You create all required subnets before installation.</td><td>You create the Flex subnet. Flex creates the remaining subnets in Flex-managed mode; you create them in customer-managed mode.</td><td>Silk creates all required VPC networks and subnets. Existing VPCs and pre-created subnets aren't supported.</td></tr><tr><td><strong>Identity and permissions</strong></td><td>IAM roles and policies</td><td>Azure managed identity (system-assigned or user-assigned, depending on deployment model)</td><td>A Deployment service account for the Marketplace deployment, then the project's Compute Engine default service account for Flex operations</td></tr><tr><td><strong>Silk images</strong></td><td>Verify access to the required Silk AMIs</td><td>Verify that the required Silk images are available before creating DataPods</td><td>Import the c.node and d.node images from the <strong>Silk-Image-Repo</strong> project into your deployment project</td></tr><tr><td><strong>Host connectivity</strong></td><td>Application hosts connect to DataPods over the Data 1 subnet</td><td>Application hosts connect to DataPods over the Data 1 subnet</td><td>Configure VPC Network Peering between your host VPC and the Silk Data 1 VPC</td></tr></tbody></table>

## Choose your cloud provider deployment path

Select the path for the environment where you'll deploy Silk.

<table data-view="cards"><thead><tr><th align="center"></th><th align="center"></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td align="center"><h2><i class="fa-aws">:aws:</i></h2></td><td align="center"><strong>Deploy Silk on AWS</strong></td><td><a href="/deploy/deploy-on-aws/aws-deployment-overview.md">Deploy on AWS</a></td></tr><tr><td align="center"><h2><i class="fa-microsoft">:microsoft:</i></h2></td><td align="center"><strong>Deploy Silk on Azure</strong></td><td><a href="/deploy/deploy-on-azure/azure-deployment-overview.md">Deploy on Azure</a></td></tr><tr><td align="center"><h2><i class="fa-google">:google:</i></h2></td><td align="center"><strong>Deploy Silk on Google Cloud</strong></td><td><a href="/deploy/deploy-on-google-cloud/google-cloud-deployment-overview.md">Deploy on Google Cloud</a></td></tr></tbody></table>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.silk.us/deploy/plan-your-deployment/plan-a-silk-deployment.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
