The new docs.trellix.com features a modernized UI and AI-powered conversational search. Content is currently available in English, with additional languages expected in November 2026. We hope you enjoy the updated experience.

NDR clusters

Prev Next

In an NDR cluster, multiple NDR appliances work cooperatively to store, detect, and retrieve data. An NDR cluster is formed when an NDR appliance is configured or connected in a way that enables distributed storage and search. This approach uses the distributed features of Elasticsearch for operation. This is similar to, but different from the NDR distributed search setup. See Distributed search.

In an NDR cluster, every appliance runs multiple Elasticsearch nodes within the box, which makes every NDR appliance a cluster. There are two types of NDR clusters: a single-box cluster and a multi-box cluster.

NDR Cluster Components

Definitions

Primary-Data Node

Collects metadata from one or more connected PX, NX, or IPS appliances, as well as AWS VPC flow logs, DNS logs, and Infoblox, and evenly distributes the data load between itself and one or more connected NDR data nodes.

Data Node

NDR appliances used to store and search metadata distributed by the connected Primary-Data Node.

Single-box cluster

A Single-Box NDR cluster is formed when an appliance has two internal data nodes that store data and allow it to be searched. One internal node functions as a Primary Data Node and the other internal node functions as a data node. The Primary Data Node collects metadata from one or more connected PX, NX, or IPS appliances, as well as AWS VPC flow logs, DNS logs, and Infoblox, and evenly distributes the data load between itself and the data node.

Multi-box cluster

A Multi-Box NDR cluster is formed when multiple NDR appliances are connected together and configured to store data and allow it to be searched. This cluster configuration significantly increases the storage and search capacity of your NDR appliances and utilizes the Primary Data Node and data node designation in a different way.

In a multi-box cluster, a single NDR appliance serves as the Primary node because it contains a primary-data node and a data node. This appliance is connected to one or more PX appliances and one or more NDR appliances, and coordinates the effort of the NDR cluster. The Primary node is necessary for the multi-box cluster operation and interaction with the NDR Web UI. The other NDR appliances that make up the multi-box cluster only contain data nodes that are used to store and search metadata.

Each NDR cluster has a name that uniquely identifies it. Each primary or data node also has a node name that uniquely identifies it.

Terms used to describe an “ NDR cluster” and a distributed search configuration are common to both deployments but have different meanings. This section focuses on NDR Clusters. Any references to distributed search terminology will be called out explicitly.

Here is an overview of the basic steps in the cluster deployment process:

  1. Choose the deployment option.

  2. Configure the primary nodes.

  3. Configure the data nodes.

  4. Start/Restart nodes.