> 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/references/glossary.md).

# Glossary

Definitions of terms you'll encounter in the docs and dashboard.

Plain-language definitions of terms used across the platform. Click any entry to expand it.

<button type="button" class="button primary" data-action="search" data-icon="magnifying-glass">Search...</button>

***

## A — F

<details>

<summary><strong>Accelerated Performance (Workload Profile)</strong></summary>

A DataPod profile that adds read cache to each c.node to improve throughput and latency for read-intensive workloads.

*Previously called Performance Acceleration or* [*ILC*](#high-performance-workload-profile) *with Read Cache*

</details>

<details>

<summary><strong>Allocated Capacity</strong></summary>

The logical amount of space used before accounting for compression and deduplication. Also called **logical capacity**.

{% hint style="info" %}
A DataPod's maximum allocated capacity is established during DataPod creation and cannot be changed afterward.
{% endhint %}

</details>

<details>

<summary><strong>Amazon Machine Image (AMI)</strong></summary>

An image used to launch an Amazon EC2 instance with a specified operating system and software configuration. Silk uses AMIs to deploy its components on AWS.

</details>

<details>

<summary><strong>Amazon Virtual Private Cloud (VPC)</strong></summary>

A logically isolated network within AWS where you deploy resources and control IP address ranges, subnets, routing, and network access.

</details>

<details>

<summary><strong>Application-Consistent Snapshot</strong></summary>

A snapshot taken while no application I/O is in progress, so the captured data represents a consistent state that the application can start from without recovery.

For databases, Silk Echo coordinates database activity and flushes pending writes so the captured data and transaction logs represent a consistent recovery point.

</details>

<details>

<summary><strong>Application Token</strong></summary>

A credential used to authenticate programmatic access to Silk Flex.

</details>

<details>

<summary><strong>AWS CloudFormation</strong></summary>

An AWS service that provisions and manages infrastructure from templates. Silk uses CloudFormation during the initial deployment of Flex on AWS.

</details>

<details>

<summary><strong>Base Volume</strong></summary>

The source volume that a snapshot was taken from. A [#snapshot-restore](#snapshot-restore "mention") writes the snapshot's data back to its base volume.

</details>

<details>

<summary><strong>Blue-Green Deployment</strong></summary>

An upgrade method that prepares a replacement environment alongside the existing environment, then redirects traffic to the replacement after validation.

</details>

<details>

<summary><strong>Call Home</strong></summary>

The Silk capability that securely sends system telemetry and diagnostic information to Silk for monitoring, analysis, and support.

</details>

<details>

<summary><strong>Capacity Policy</strong></summary>

A policy that controls or governs capacity-related behavior for DataPod resources.

</details>

<details>

<summary><strong>Capacity Stress</strong></summary>

A capacity state that begins at 96% usage under the default capacity policy. In this state, Silk sends a critical notification, prevents new snapshots and views from being created, and deletes existing snapshots and views. Mapped views and snapshots that are protected from auto-deletion are retained.

Capacity thresholds are configurable for volume groups. See also [#low-free-capacity](#low-free-capacity "mention").

</details>

<details>

<summary><strong>Capacity Utilization</strong></summary>

The amount or percentage of available storage capacity currently in use.

</details>

<details>

<summary><strong>Cascade Deletion</strong></summary>

An administrative operation that deletes an Echo resource and its dependent Echo resources together.

</details>

<details>

<summary><strong>Cascading Replication</strong></summary>

A replication configuration in which a target volume group is used as the source for another replication session, allowing data from one source volume group to be replicated to multiple target systems.

</details>

<details>

<summary><strong>Challenge Handshake Authentication Protocol (CHAP)</strong></summary>

An authentication protocol that verifies the identity of an iSCSI initiator before allowing it to connect to a DataPod target. CHAP uses a shared username and secret to help prevent unauthorized host access.

</details>

<details>

<summary><strong>CIDR Notation</strong></summary>

A format for expressing an IP address range using an address and a prefix length, such as `10.0.0.0/28`. A smaller prefix number represents a larger address range.

</details>

<details>

<summary><strong>Clone</strong></summary>

A writable copy of a specific volume, created from a snapshot. A clone can be created as a [#thin-clone](#thin-clone "mention") or a [#full-clone](#full-clone "mention"), depending on how its data and capacity are managed.

Unlike a [#view](#view "mention"), which represents an entire volume group, a clone copies a single volume.

</details>

<details>

<summary><strong>Cluster</strong></summary>

A Silk environment managed by a Flex instance and containing one or more DataPods.

</details>

<details>

<summary><strong>Cluster Configuration File</strong></summary>

An environment-specific file supplied by Silk that contains configuration inputs required to create a Silk cluster. The file is based on the deployment's pre-installation planning and cannot be reused for another environment.

</details>

<details>

<summary><strong>Cluster Size</strong></summary>

A scale selected when deploying a Silk cluster that determines the maximum numbers of DataPods, c.nodes, and m.nodes the cluster can support.

{% hint style="warning" %}
Cluster size can't be changed after the cluster is created.
{% endhint %}

</details>

<details>

<summary><strong>Compression</strong></summary>

A data reduction method that encodes data so it consumes less physical storage. Silk performs compression inline and in real time as data is written to a DataPod.

</details>

<details>

<summary><strong>Compute Node (c.node)</strong></summary>

A DataPod resource that delivers high-performance block data services to connected hosts.

</details>

<details>

<summary><strong>Copy Data Management (CDM)</strong></summary>

The practice of creating, managing, refreshing, and retiring data copies for recovery, development, testing, analytics, and other secondary workloads.

</details>

<details>

<summary><strong>Crash-Consistent Snapshot</strong></summary>

A snapshot taken without pausing application I/O. It captures data as it exists on the volumes at that moment, including writes still in progress at the application level, so an application or database that starts from it may need to perform recovery.

Crash-consistent is the default snapshot type.

</details>

<details>

<summary><strong>Cross-Volume-Group Snapshot</strong></summary>

A snapshot that captures multiple volume groups at the same point in time. Use it when an application's data spans more than one volume group.

</details>

<details>

<summary><strong>Customer-Managed Mode (UMI)</strong></summary>

An Azure deployment model in which Silk Flex uses a user-assigned managed identity (UMI) and customer-prepared networking and permissions. Your organization creates and manages the required subnets, network security groups, and identity permissions.

</details>

<details>

<summary><strong>Data Mobility</strong></summary>

The movement or reuse of data across DataPods, environments, or workloads.

</details>

<details>

<summary><strong>Data Node (d.node)</strong></summary>

A DataPod resource that provides the persistent storage capacity managed through media nodes.

</details>

<details>

<summary><strong>Data Protection</strong></summary>

The processes and technologies used to preserve data availability, integrity, and recoverability.

</details>

<details>

<summary><strong>Data Reduction</strong></summary>

A collection of technologies that reduces physical storage consumption. Silk performs data reduction inline as data is written, using global deduplication, compression, and zero elimination.

[#thin-provisioning](#thin-provisioning "mention") further improves capacity efficiency by allocating storage on demand.

</details>

<details>

<summary><strong>Data Service</strong></summary>

A storage capability provided by a DataPod, such as snapshots, clones, replication, thin provisioning, or data reduction.

</details>

<details>

<summary><strong>Database Block Sizing</strong></summary>

The process of determining how much data each database block can store. Silk dynamically adapts block sizes to application workloads to support consistent performance across different workload types.

</details>

<details>

<summary><strong>Database Lineage</strong></summary>

The traceable relationship between a source database, its DB Snapshots, and the Echo Databases created through successive cloning operations.

</details>

<details>

<summary><strong>Database Refresh</strong></summary>

An Echo operation that updates an existing database copy using data from a newer snapshot.

</details>

<details>

<summary><strong>DB Snapshot</strong></summary>

A point-in-time representation of a database created through coordination between Echo, the database host, and the underlying DataPod.

You can create an [#application-consistent-snapshot](#application-consistent-snapshot "mention") or a [#crash-consistent-snapshot](#crash-consistent-snapshot "mention").

</details>

<details>

<summary><strong>Deduplication</strong></summary>

A data reduction method that stores a single instance of duplicate data to reduce physical storage consumption. Silk performs deduplication globally and inline as data is written across a DataPod.

</details>

<details>

<summary><strong>Deployment Profile</strong></summary>

A predefined DataPod configuration that determines the available c.node and m.node options, capacity ranges, performance characteristics, and scaling options.

</details>

<details>

<summary><strong>Echo Constellation</strong></summary>

A visual workspace in Echo for exploring hosts, databases, DB Snapshots, Echo Databases, and their relationships.

</details>

<details>

<summary><strong>Echo Database (EchoDB)</strong></summary>

A full, writable database provisioned by Echo from a database snapshot for use on a supported target host.

</details>

<details>

<summary><strong>Echo of Echo</strong></summary>

An Echo capability for creating an Echo Database from another Echo Database, producing a multi-generation database lineage.

</details>

<details>

<summary><strong>Echo Host Agent</strong></summary>

Lightweight software installed on a supported database host that enables Echo to perform host- and database-level operations.

</details>

<details>

<summary><strong>Echo Source Database</strong></summary>

A database registered with Echo that allows Echo to manage database-aware snapshot and copy workflows.

</details>

<details>

<summary><strong>Elastic Network Interface (ENI)</strong></summary>

A virtual network interface in an AWS VPC. VPC interface endpoints create ENIs that consume IP addresses in the subnets where they are placed.

</details>

<details>

<summary><strong>Entry-level DataPod</strong></summary>

A smaller DataPod type available for supported Azure Managed Disk (Azure Pv2) deployments. It provides reduced capacity and scaling options for development, testing, proofs of concept, and smaller production workloads.

</details>

<details>

<summary><strong>Flex-Managed Mode (SMI)</strong></summary>

An Azure deployment model in which Silk Flex uses a system-assigned managed identity (SMI) and manages the required cluster resources and permissions. Your organization prepares the VNet and Flex subnet before installation.

</details>

<details>

<summary><strong>Full Clone</strong></summary>

An independent, fully physical copy of source data.

</details>

***

## G — P

<details>

<summary><strong>High Performance (Workload Profile)</strong></summary>

The default DataPod profile for most production workloads. It provides balanced performance and scalability and supports the standard production allocated capacity range.

*Previously called Increased Logical Capacity or ILC.*

</details>

<details>

<summary><strong>Host</strong></summary>

An application server, database server, or virtual machine that connects to a Silk DataPod to access mapped storage. In the SDP Management Console, a host is identified by its iSCSI Qualified Name (IQN).

</details>

<details>

<summary><strong>Host Group</strong></summary>

A collection of hosts that you can map to a volume together. A host can belong to one host group.

</details>

<details>

<summary><strong>Independent Volume Group</strong></summary>

A default volume group for independently managed deduplicated or non-deduplicated volumes.

Independent volume groups can't be modified or deleted and do not support snapshots or replication.

</details>

<details>

<summary><strong>IOPS</strong></summary>

Input/output operations per second. A measure of how many read and write operations a storage system completes each second.

</details>

<details>

<summary><strong>iSCSI Qualified Name (IQN)</strong></summary>

A unique name that identifies an iSCSI initiator or target.

</details>

<details>

<summary><strong>Latency</strong></summary>

The time required for a system to complete an operation or respond to a request. Storage latency is typically measured in milliseconds.

</details>

<details>

<summary><strong>Limited Capacity (Workload Profile)</strong></summary>

A DataPod profile that supports lower logical capacity configurations for smaller environments.

</details>

<details>

<summary><strong>Locked Snapshot</strong></summary>

A snapshot that can't be deleted until its dependencies are removed or it's unlocked. A snapshot can be locked because of dependent views or clones, replication dependencies, or retention policy settings.

In the SDP Management Console, locked snapshots are marked with a lock icon.

</details>

<details>

<summary><strong>Low Free Capacity</strong></summary>

A capacity state that begins at 90% usage under the default capacity policy. In this state, Silk sends an error notification and deletes older snapshots, except snapshots protected from auto-deletion. Views are retained.

Capacity thresholds are configurable for volume groups. See also [#capacity-stress](#capacity-stress "mention").

</details>

<details>

<summary><strong>Managed Disk Deployment</strong></summary>

A Silk deployment architecture that uses cloud-managed storage for d.nodes instead of storage attached locally to the m.node virtual machine.

This deployment architecture is available for Azure Premium SSD v2 configurations.

</details>

<details>

<summary><strong>Managed Identity</strong></summary>

An Azure identity that allows a resource like Silk Flex to authenticate with Azure services, without customer-managed credentials.

Flex supports **system-assigned managed identities (SMI)**, whose lifecycle is tied to the associated Azure resource, and **user-assigned managed identities (UMI)**, which are created and managed independently.

</details>

<details>

<summary><strong>Media node (m.node)</strong></summary>

A media node that provides persistent capacity and data durability for a DataPod by organizing a group of d.nodes into a protected logical storage resource.

</details>

<details>

<summary><strong>m.node Live Migration</strong></summary>

A supported scaling operation that moves data from an existing m.node to a replacement m.node while the DataPod remains online.

</details>

<details>

<summary><strong>Module-Based Non-Disruptive Upgrade (mNDU)</strong></summary>

A targeted upgrade that updates selected Silk management components independently without interrupting application I/O.

</details>

<details>

<summary><strong>NAT Gateway</strong></summary>

An AWS networking service that allows resources in private subnets to initiate outbound connections without accepting unsolicited inbound connections.

</details>

<details>

<summary><strong>Non-Disruptive Upgrade (NDU)</strong></summary>

A Silk software upgrade that maintains application access to data while platform components are updated.

Management services may be temporarily unavailable during the upgrade.

</details>

<details>

<summary><strong>Orchestration</strong></summary>

The automated coordination and management of resources and operations across an environment.

</details>

<details>

<summary><strong>Pre-Installation Site Survey</strong></summary>

A planning artifact completed with the Silk deployment team to document the configuration, networking, and resource requirements for a deployment.

</details>

<details>

<summary><strong>Physical Capacity</strong></summary>

The amount of underlying storage consumed after applicable compression and deduplication savings.

</details>

<details>

<summary><strong>Provisioned Capacity</strong></summary>

The total configured storage capacity assigned to volumes, clones, views, and snapshots, regardless of how much data has been written.

Provisioned capacity represents the storage made available to workloads and is used for database sizing and Silk licensing.

</details>

***

## R — Z

<details>

<summary><strong>Recovery Point Objective (RPO)</strong></summary>

The maximum acceptable amount of data loss after a disruption, expressed as a length of time. In Silk replication, the RPO determines how often the replication engine transfers snapshots to the target DataPod.

</details>

<details>

<summary><strong>Redirect-on-Write (RoW)</strong></summary>

The snapshot method Silk uses to preserve point-in-time data. When a host writes new data, the DataPod writes it to a new location and keeps the original blocks for existing snapshots, so a snapshot consumes additional capacity only as source data changes.

</details>

<details>

<summary><strong>Replication</strong></summary>

The transfer of data between DataPods or databases to support disaster recovery and data mobility.

DataPod replication is asynchronous, while database-level replication may be synchronous or asynchronous.

</details>

<details>

<summary><strong>Replication Peer</strong></summary>

A connection between two DataPods that allows data to be replicated from one DataPod to the other.

</details>

<details>

<summary><strong>Replication Session</strong></summary>

A configured relationship that replicates a volume group from a source DataPod to a target DataPod according to defined recovery point objective (RPO) and retention settings.

</details>

<details>

<summary><strong>Retention Policy</strong></summary>

A policy that defines how long a volume group keeps its snapshots, based on a retention period or a number of snapshots to keep. Silk includes several default retention policies, and you can create custom policies.

Snapshots with dependent views or clones are locked and can't be deleted, regardless of their retention policy.

</details>

<details>

<summary><strong>SDP Management Console</strong></summary>

The browser-based management interface used to administer the storage resources and data services within an individual DataPod.

</details>

<details>

<summary><strong>Secondary Workload</strong></summary>

A workload that uses production or production-derived data for a purpose other than running the primary production application, such as development, testing, reporting, analytics, or AI.

</details>

<details>

<summary><strong>SecureLink</strong></summary>

An optional remote support connection that lets Silk Support access a DataPod for troubleshooting and upgrades when enabled.

</details>

<details>

<summary><strong>Service Account</strong></summary>

A Google Cloud identity that software uses to authenticate and perform authorized operations. Silk uses service accounts to deploy Flex and manage cloud resources.

Silk uses service accounts for two purposes:

* **Deployment service account:** Used to provision Flex through Google Cloud Marketplace.
* **Flex service account:** Used by Flex after deployment to create clusters, provision DataPods, and manage Silk cloud resources, including upgrades and deletion. By default, Flex uses the project's Compute Engine default service account. You can configure a custom account if needed.

</details>

<details>

<summary><strong>Silk Clarity</strong></summary>

A cloud-based monitoring and analytics product that collects Silk platform telemetry for health, utilization, and performance analysis.

</details>

<details>

<summary><strong>Silk Cloud Data Platform</strong></summary>

A software-defined data platform that transforms cloud compute and storage resources into a shared space-efficient, high-performance data layer for databases and applications.

</details>

<details>

<summary><strong>Silk DataPod</strong></summary>

A virtualized storage environment that provides block storage and data services using native cloud compute and storage resources.

</details>

<details>

<summary><strong>Silk Echo</strong></summary>

A copy data management product that automates the creation and management of full-performance database copies using DataPod data services.

</details>

<details>

<summary><strong>Silk Flex</strong></summary>

The orchestration and management interface used to deploy, view, scale, and manage Silk DataPods and their supporting cloud resources.

</details>

<details>

<summary><strong>Silk Host Agent</strong></summary>

Software installed on a database host that discovers databases and coordinates operations between a database host and Echo.

</details>

<details>

<summary><strong>Silk Network</strong></summary>

Silk uses six dedicated networks to separate management, application data, replication, and internal cluster traffic. The following table describes each network's role:

<table><thead><tr><th width="144.073974609375">Network</th><th>Description</th></tr></thead><tbody><tr><td><strong>Flex</strong></td><td>Hosts Silk Flex and provides connectivity for its management and API services.</td></tr><tr><td><strong>External Management</strong></td><td>Supports management traffic and communication between Silk components.</td></tr><tr><td><strong>Data 1</strong></td><td>Carries iSCSI traffic between application hosts and Silk DataPods. Hosts connect to <code>dataport01</code> on each c.node.</td></tr><tr><td><strong>Data 2</strong></td><td>Carries replication traffic between Silk DataPods. Each c.node uses <code>dataport02</code> for this traffic.</td></tr><tr><td><strong>Internal 1 and Internal 2</strong></td><td>Support internal cluster communication.</td></tr></tbody></table>

</details>

<details>

<summary><strong>Snapshot</strong></summary>

An instantaneous, point-in-time representation of data that does not initially create a full physical copy. Silk creates snapshots at the volume group level, capturing all volumes and clones in the group together.

Snapshots can't be mapped to hosts directly. To access snapshot data, create a [#view](#view "mention") or a [#clone](#clone "mention"). For details, see [Snapshots](/learn/data-management-concepts/snapshots.md).

</details>

<details>

<summary><strong>Snapshot Restore</strong></summary>

An operation that writes a snapshot's data back to its [#base-volume](#base-volume "mention"), overwriting the volume's current data.

</details>

<details>

<summary><strong>Snapshot Scheduler</strong></summary>

A configuration that automatically creates snapshots at a defined interval. You can assign one scheduler to multiple volume groups, and a volume group can use multiple schedulers.

</details>

<details>

<summary><strong>Standard DataPod</strong></summary>

A DataPod type that provides the full supported Silk feature set and the greatest configuration and scaling flexibility.

</details>

<details>

<summary><strong>Storage Type</strong></summary>

A DataPod configuration that determines the m.node and d.node architecture used to provide physical capacity, such as VM Local Disks or Azure Managed Disks.

</details>

<details>

<summary><strong>System Event</strong></summary>

A recorded occurrence or change within a system that can provide information about its health, status, or operation.

</details>

<details>

<summary><strong>Thin Clone</strong></summary>

A writable data copy that shares unchanged data with its source and consumes additional capacity only as changes are written.

</details>

<details>

<summary><strong>Thin Provisioning</strong></summary>

A method of allocating physical storage capacity as data is written rather than reserving the full volume size in advance.

</details>

<details>

<summary><strong>Throughput</strong></summary>

The amount of data transferred by a system during a given period, typically measured in megabytes or gigabytes per second.

</details>

<details>

<summary><strong>View</strong></summary>

A writable representation of a volume group snapshot that can be mapped to hosts. A view exposes the data from all volumes in the snapshot and consumes additional capacity only as its data changes.

Changes made through a view don't affect its source snapshot.

</details>

<details>

<summary><strong>VM Local Disk Deployment</strong></summary>

A Silk deployment architecture in which d.node storage is provided by disks attached locally to the m.node virtual machines.

</details>

<details>

<summary><strong>Volume</strong></summary>

A block storage resource provisioned from a DataPod and presented to a connected host.

</details>

<details>

<summary><strong>Volume Group (VG)</strong></summary>

A logical container that organizes related volumes within a DataPod. Volume groups let you manage capacity and apply data protection settings—including snapshots, retention policies, and replication—across multiple volumes.

</details>

<details>

<summary><strong>Volume Mapping</strong></summary>

An association between a volume and a host or host group that lets the DataPod present the volume to the associated hosts.

</details>

<details>

<summary><strong>VPC Endpoint</strong></summary>

An AWS networking resource that provides private connectivity from a VPC to a supported service without requiring internet access.

Interface endpoints use network interfaces in selected subnets, while gateway endpoints use route tables.

</details>

<details>

<summary><strong>VPC Network Peering</strong></summary>

A Google Cloud networking connection that enables communication between VPC networks using internal IP addresses. In a Silk deployment, it connects the application host VPC to the Silk Data 1 VPC for iSCSI traffic.

</details>

<details>

<summary><strong>Zero Elimination</strong></summary>

A data reduction method that avoids physically storing blocks containing only zeros.

</details>

***

{% hint style="info" %}
See a term missing? Leave us feedback. We update this glossary regularly.
{% endhint %}


---

# 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/references/glossary.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.
