> 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/learn/products-and-core-concepts/datapod-architecture/clusters-and-datapods.md).

# Clusters and DataPods

Learn how clusters organize your Silk environment, how DataPods deliver data services, and how c.nodes, m.nodes, and d.nodes work together to provide scalable performance and capacity.

Silk organizes cloud resources into clusters and DataPods. Together, these layers define how you deploy, scale, and manage data services across your Silk environment.

A **cluster** provides the foundation for the environment, including shared networking, resource limits, and global settings. Within the cluster, one or more **DataPods** provide the data services used by your applications and databases.&#x20;

```mermaid
flowchart TB
    F["Silk Flex<br/>Management and orchestration"]
    C["Cluster<br/>Shared configuration and networking"]
    DP["Silk DataPod<br/>One or more per cluster"]
    CN["c.nodes<br/>Performance and data services"]
    MN["m.nodes<br/>Capacity and durability"]
    DN["d.nodes<br/>Persistent data storage"]

    F -->|Manages| C
    C -->|Contains| DP
    DP -->|Includes| CN
    DP -->|Includes| MN
    MN -->|Organizes and protects| DN

    classDef management fill:#FCE8F1,stroke:#B62468,color:#40102A
    classDef datapod fill:#F5D3E3,stroke:#B62468,color:#40102A
    classDef component fill:#FFF7FA,stroke:#D36A9B,color:#40102A

    class F,C management
    class DP datapod
    class CN,MN,DN component
```

## How clusters and DataPods fit together

Each Silk Flex instance manages a single cluster. That cluster can contain one or more DataPods, depending on its size and the requirements of your environment.&#x20;

<table><thead><tr><th width="219.333251953125">Layer</th><th>Purpose</th></tr></thead><tbody><tr><td><strong>Silk Flex</strong></td><td>Deploys and manages the cluster and its DataPods.</td></tr><tr><td><strong>Cluster</strong></td><td>Defines the shared foundation for the Silk environment, including networking, global settings, and the maximum number of supported resources.</td></tr><tr><td><strong>DataPod</strong></td><td>Delivers high-performance data services to connected applications and databases.</td></tr><tr><td><strong>c.nodes</strong></td><td>Provide the DataPod's performance and data-processing layer.</td></tr><tr><td><strong>m.nodes and d.nodes</strong></td><td>Provide the DataPod's persistent capacity and data-durability layer.</td></tr></tbody></table>

This structure allows an organization to operate multiple DataPods within the same Silk environment. For example, you might use separate DataPods for different application groups, performance requirements, environments, or administrative boundaries.

{% hint style="info" %}
A DataPod isn't the same as a cluster. The cluster establishes the environment in which DataPods are deployed.&#x20;

Applications connect to storage resources provisioned from individual DataPods.
{% endhint %}

## Clusters

A cluster is the top-level grouping of Silk resources managed through a Flex instance. It provides the shared configuration and network foundation required to deploy and operate DataPods.

When a cluster is created, you define settings that apply across the environment, including:

* Cluster and organization information
* Network address space
* Management, data, replication, and internal communications
* Callhome configuration
* NTP and time zone settings
* Maximum number of Silk resources the environment can support

Clusters are available in different sizes. The selected cluster size determines the network address space required by Silk and the maximum number of resources your organization can deploy.&#x20;

{% hint style="warning" %}
**Cluster size cannot be changed**

The cluster size cannot be changed after creation. Plan ahead for the potential growth of your DataPods before deployment.&#x20;

If an existing cluster no longer supports your requirements, replacing it with a differently sized cluster requires moving its dependent DataPods first.&#x20;
{% endhint %}

### Cluster components

A cluster contains one or more DataPods and the cloud resources required to support them. It also provides the networks used for:

* Flex and platform management
* Communication with connected application hosts
* Replication between DataPods
* Internal communication between DataPod components

The specific cloud resources and network implementation vary between AWS, Azure, and Google Cloud. Refer to the deployment documentation for your cloud provider when planning or creating a cluster.&#x20;

<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"><h3>AWS</h3></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"><h3>Microsoft Azure</h3></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"><h3>Google Cloud</h3></td><td><a href="/deploy/deploy-on-google-cloud/google-cloud-deployment-overview.md">Deploy on Google Cloud</a></td></tr></tbody></table>

## DataPods

Silk DataPods deliver scalable, high-performance data services for applications and databases. Within a Silk environment, you deploy one or more individual DataPods.

Each DataPod combines cloud infrastructure with Silk data services to provide:

* **High-performance shared storage:** Delivers shared block storage for application and database workloads.
* **Flexible resource scaling:** Allows performance and capacity to scale independently as workload requirements change.
* **Efficient data management:** Uses thin provisioning and data reduction to improve storage efficiency.
* **Data protection and administration:** Supports snapshots, writable copies, replication, disaster recovery, and centralized management of hosts and volumes.&#x20;

### DataPod architecture

Every DataPod uses three distinct node types: **c.nodes**, **m.nodes**, and **d.nodes**. Each node type supports a different part of the data path:

* **c.nodes:** Performance and data services
* **m.nodes:** Capacity and durability
* **d.nodes:** Persistent data storage

#### c.nodes

Compute nodes, or **c.nodes**, provide the processing layer for a DataPod. They run as cloud virtual machines and deliver block data services to connected applications and databases.&#x20;

c.nodes are responsible for activities like:

* Receiving I/O from connected hosts
* Processing reads and writes
* Distributing workloads across the DataPod
* Applying data services and optimizations
* Communicating with the DataPod's capacity layer

A DataPod uses multiple c.nodes to provide an active-active architecture and maintain service availability. Workloads are distributed across the available c.nodes rather than being tied to a single controller.

Adding c.nodes increases the processing resources available to the DataPod. This allows you to scale performance without automatically adding storage capacity.&#x20;

#### m.nodes

Media nodes, or **m.nodes**, provide the persistent capacity layer for a DataPod. An m.node acts as a virtual media or drive shelf and organizes a group of d.nodes into a protected storage resource.

m.nodes provide the capacity used by volumes, snapshots, and copies. Depending on the cloud provider and deployment type, an m.node can be backed by cloud virtual machines with local media or by cloud-native managed disks.

Adding m.nodes increases the physical capacity available to a DataPod independently from its c.nodes. For supported configurations, m.node live migration can also move data from an existing m.node to a replacement m.node while the DataPod remains online. This allows you to change the underlying capacity resources as infrastructure and cost requirements evolve.&#x20;

#### **d.nodes**

Data nodes, or **d.nodes**, are the individual storage components that hold and manage persistent data. Multiple d.nodes are grouped together to form an m.node.

Silk distributes data and protection information across the d.nodes in the DataPod. This distributed design provides data durability while allowing the capacity layer to use the underlying cloud infrastructure efficiently.&#x20;

You normally manage capacity at the m.node level rather than administering individual d.nodes. However, you can view the status of both m.nodes and d.nodes through the SDP Management Console.&#x20;

{% columns %}
{% column %}

### How data flows through a DataPod

{% endcolumn %}

{% column %}
{% stepper %}
{% step %}

#### A connected host sends I/O to the DataPod.

{% endstep %}

{% step %}

#### The c.nodes process and optimize data.

{% endstep %}

{% step %}

#### The DataPod distributes the data across its capacity layer.

{% endstep %}

{% step %}

#### The m.nodes organize and protect the data across their underlying d.nodes.

{% endstep %}
{% endstepper %}
{% endcolumn %}
{% endcolumns %}

When an application reads data, the c.nodes retrieve the required data from the capacity layer and return it to the host.

Volumes are accessible through the DataPod rather than being permanently associated with a particular c.node or m.node. This shared architecture allows Silk to redistribute work and use additional resources as the DataPod grows.

### Scale performance and capacity independently

A central characteristic of the DataPod architecture is the separation of performance and capacity.

* **To increase processing performance**, add c.nodes.
* **To increase physical capacity**, add or migrate m.nodes.
* **To reduce resources**, remove supported c.nodes or m.nodes through the applicable scaling workflow.

This separation lets you adjust a DataPod for the requirement that has changed. A workload that needs more throughput doesn't necessarily require more storage, while a growing dataset doesn't necessarily require additional compute resources.&#x20;

When capacity resources change, Silk handles the distribution of data across the DataPod. Hosts continue to access storage through the same DataPod rather than being reconfigured for each underlying resource change.

Scaling options and limits depend on the cloud provider, deployment type, DataPod type, and selected deployment profile. Review the current platform limits before changing a DataPod.&#x20;

<table data-view="cards"><thead><tr><th align="center"></th><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"><h3><i class="fa-arrow-up-right-and-arrow-down-left-from-center" style="color:$primary;">:arrow-up-right-and-arrow-down-left-from-center:</i></h3></td><td align="center"><strong>Scale a DataPod</strong></td><td align="center">Add or remove resources as your performance and capacity requirements change.</td><td></td></tr><tr><td align="center"><h2><i class="fa-gauge-max" style="color:$primary;">:gauge-max:</i></h2></td><td align="center"><strong>Review DataPod limits</strong></td><td align="center">Check supported configurations and scaling limits before making changes.</td><td><a href="/learn/references/silk-platform-configuration-limits.md">Silk Platform Configuration Limits</a></td></tr></tbody></table>

## Manage clusters and DataPods

Silk provides different management experiences for the cluster and for the storage resources inside each DataPod:

<table><thead><tr><th width="234.99993896484375">Interface</th><th>Use it to</th></tr></thead><tbody><tr><td><strong>Silk Flex</strong></td><td>View and manage the cluster, deploy and scale DataPods, manage cloud resources, and access environment-level operations.</td></tr><tr><td><strong>SDP Management Console</strong></td><td>Manage an individual DataPod, including hosts, volumes, volume groups, snapshots, copies, replication, capacity policies, events, and components.</td></tr><tr><td><strong>Silk Clarity</strong></td><td>Monitor performance, capacity, data protection, and operational trends across the Silk environment.</td></tr></tbody></table>

From Flex, you can review DataPod health, configuration, capacity, bandwidth, IOPS, and latency. You can then open the SDP Management Console for an individual DataPod when you need to provision or manage its storage resources.&#x20;

## Next steps

* **Create a DataPod.** Select a deployment profile and provision a DataPod through Silk Flex.
* [**Explore Silk Flex**](/learn/products-and-core-concepts/silk-product-fundamentals/silk-flex-overview.md)**.** Learn how to navigate and manage your cluster and DataPods.
* [**Explore the SDP Management Console.**](/learn/products-and-core-concepts/datapod-architecture/sdp-management-console-overview.md) Find the tools used to administer storage resources within a DataPod.
* [**Connect hosts and provision storage**](/deploy/connect-hosts/host-connectivity.md)**.** Make DataPod volumes available to your applications and databases.
* [**Monitor your environment**](/manage/monitor-and-automate-silk/monitor-your-silk-environment.md)**.** Track health, performance, and capacity through Silk Clarity.


---

# 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/learn/products-and-core-concepts/datapod-architecture/clusters-and-datapods.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.
