Elastic Stack System Requirements

Install Elastic Stack Components

The first step of the Environment Watch and/or Data Grid Audit setup involves installing required third-party software from the Elastic Stack. The following Elastic components are used for these two products:

Elastic Component	Environment Watch	Data Grid Audit
Elasticsearch	Required	Required
Kibana	Required	Recommended but optional
APM Server	Required	Not used

The following actions are performed:

• Set up and verify the Elasticsearch cluster
• Set up Kibana and/or APM Server, depending on the product configuration

This step covers configuring a shared Elasticsearch cluster to be used for both Environment Watch and Data Grid Audit.

Before proceeding, the following section introduces key Elastic Stack concepts.

Elastic Key Concepts

Elasticsearch

See the port diagram for network requirements.

Elasticsearch is a distributed search and analytics engine, scalable data store, and vector database built on Apache Lucene. It's optimized for speed and relevance on production-scale workloads.

Elasticsearch Cluster

Elasticsearch is built on a distributed architecture made up of many servers or nodes. An Elasticsearch cluster is a group of one or more node instances that are connected together. Clustering enables Elasticsearch nodes to work together to ensure high availability when one or more nodes are down. An Elasticsearch cluster can continue operating normally if some of its nodes are unavailable or disconnected, as long as there are enough well-connected nodes to ensure high resilience and improved search performance.

See below for more information on Elasticsearch cluster configuration and high availability options:

• Cluster configuration
• Set up a cluster for high availability

Elasticsearch Nodes

An Elasticsearch node is a single server that is a part of a cluster. If a single node of Elasticsearch is running, then there is a cluster of one node. A node can have one or many roles that define the role(s) it plays within the cluster. The node role(s) is defined within the elasticsearch.yml . If roles are not set, the node is assigned to most available roles.

See here for more information on Elasticsearch node settings and roles.

The two key roles for an Environment Watch and/or Data Grid Audit cluster are master and data.

• Master Node: controls the Elasticsearch cluster and is responsible for all cluster-wide operations like creating/deleting an index and adding/removing nodes.
• Data Node: stores data and executes data-related operations such as search and aggregation.

Every Elasticsearch cluster requires at least one node designated master and data.

Kibana

See the port diagram for network requirements.

Kibana is a user interface that lets users visualize Elasticsearch data and navigate the Elastic stack. See below for more information on Kibana.

• What is Kibana?
• Kibana key concepts

APM Server

See the port diagram for network requirements.

The APM Server provides a fully Open Telemetry compliant telemetry backend to periodically receive log, metric, and trace data from each monitored server.

See here for more information about Elastic's APM Server.

System Requirements and Licensing

Setting up the Elastic stack components will require installing Elastic software on one or more servers. This installation guide covers configuring a shared Elasticsearch cluster to use for both Environment Watch and Data Grid Audit. The Hardware Recommendations by Environment Size section below provides hardware guidance for different deployment scenarios and reference environments.

Small environment size used for Environment Watch and Data Grid Audit_

Software Requirements

Any server being used to host Elastic components requires:

• Operating System - See here for supported operating systems.

Hardware Recommendations by Environment Size

The number of servers and hardware specifications needed to host the Elastic components will vary depending on the size of the Relativity instance and whether the cluster is intended for Environment Watch, Data Grid Audit, or both. Below are recommendations based on four Relativity Server environment sizes. These are only recommendations. The node counts and role blends can be adjusted for the environment based on observed and desired performance and reliability needs.

A few other key notes and reminders:

• Tuning for speed - Review Elastic's guidance on how to tune the environment for speed here.

Environment Size

Environment Size	Web Servers	Agent Servers	Worker Servers	SQL Distributed Servers
Development	1	2	1	1
Small	2	10	2	2
Medium	8	20	6	6
Large	12	40	10	12
X-Large	24	80	10	16

Elastic Stack Infrastructure Recommendations

Environment Size	DG/Audit Data Nodes	Environment Watch Data Nodes	APM Servers	Kibana Servers
Development	1 / 500 GB	1 / 1 TB	1	1
Small	1 / 1 TB	1 / 2 TB	1	1
Medium	2 / 2 TB	2 / 3 TB	1	1
Large	5 / 16 TB	3 / 8 TB	2	2
X-Large	10 / 32 TB	5 / 16 TB	3	3

• Separate Elastic clusters is supported when using both Audit/Environment Watch but not required
• Application Performance Monitoring(APM)/Kibana servers can be load balanced
• Each Elasticsearch node should have at least 4 vCPU and 32 GB RAM.
• A single Data node can be used for both Audit and Environment Watch in Development environments.

Licensing

Both Environment Watch and Data Grid Audit require only a free and open ("Basic") Elastic license. By default, new installations have a Basic license that never expires. If you would like to utilize additional Elastic features from the Platinum or Enterprise subscription, you will need to purchase the license separately.

If you have used Elasticsearch for the optional Data Grid Audit feature on Relativity Server prior to April 2025, you would have been using a Platinum license key provided by Relativity. Effective with Server 2024 Patch 1, the Platinum license is no longer required for Data Grid Audit and Relativity will not provide a Platinum license for any new deployments of Data Grid Audit. All existing Data Grid Audit customers will have until early 2026 to adopt Relativity Server 2024 and update to a Basic Elastic license.

Installing the Elastic stack components

Before you start

1. Plan your cluster - based on the size of your environment, establish a game plan for how many servers/nodes you intend to use and the role of each node within the cluster. We recommend using a Development environment to test drive Environment Watch in order to get up and running as quickly as possible.

2. Review the required ports for Elastic components - See the port diagram for network requirements and ensure all necessary ports are open between servers hosting Elasticsearch, Kibana, and APM Server.

3. Windows should be updated to support long paths - This is recommended because the Elastic Stack components have dense file paths when fully extracted. In many cases, the product appears to function properly but certain features may fail because files may be missing or cannot be accessed because the path exceeds the max Windows character length. Run "gpedit.msc" to navigate into Local Group Policy Editor | Computer Configuration | Administrative Template | System | Filesystem. Double click on Enable Win32 Long path. Select the Enable radio button, click Apply, and then click OK.

   https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation?tabs=registry#enable-long-paths-in-windows-10-version-1607-and-later

4. Verify the minimum supported version of Elastic - All Relativity products have been certified with Elastic Stack 8.x or 9.x, though Data Grid Audit alone is supported on 7.17.x. If you intend to use a single cluster for both, you must install the same version of Elasticsearch on all nodes in the cluster, and that version must be the higher of the minimum versions for Environment Watch and Data Grid Audit, if different.

4. At least the minimum Relativity major version and patch specified in the Environment Watch bundle you intend to deploy is installed on all servers in the environment. See the Elastic Stack Installation Overview for the minimum version required.

5. At least the minimum supported version of Windows Server for the major version of Relativity installed in your environment is installed on each target server in your cluster (see here).