Plan Your Agent Deployment

Plan your Cortex XDR agent software deployment carefully.
You typically deploy
XDR agent software to endpoints across a network after an initial proof of concept (POC), which simulates your corporate production environment. During the POC or deployment stage, you analyze security events to determine which are triggered by malicious activity and which are due to legitimate processes behaving in a risky or incorrect manner. You also simulate the number and types of endpoints, the user profiles, and the types of applications that run on the endpoints in your organization and, according to these factors, you define, test, and adjust the security policy for your organization.
The goal of this multi-step process is to provide maximum protection to the organization without interfering with legitimate workflows.
After the successful completion of the initial POC, we recommend a multi-step implementation in the corporate production environment for the following reasons:
  • The POC doesn't always reflect all the variables that exist in your production environment.
  • There is a rare chance that the
    XDR agent will affect business applications, which can reveal vulnerabilities in the software as a prevented attack.
  • During the POC, it is much easier to isolate issues that appear and provide a solution before full implementation in a large environment where issues could affect a large number of users.
A multi-step deployment approach ensures a smooth implementation and deployment of the
solution throughout your network. Use the following steps for better support and control over the added protection.
0. Calculate the bandwidth required to support the number of agents you plan to deploy.
as needed
For every 100,000 agents, you will need to allocate 120Mbps of bandwidth. The bandwidth requirement scales linearly. For example, to support 300,000 agents, plan to allocate 360Mbps of bandwidth (three times the amount required for 100,000 agents).
1. Install Cortex XDR on endpoints.
1 week
Install the
XDR agent on a small number of endpoints (3 to 10).
Test normal behavior of the
XDR agents (injection and policy) and confirm that there is no change in the user experience.
2. Expand the
2 weeks
Gradually expand agent distribution to larger groups that have similar attributes (hardware, software, and users). At the end of two weeks you can have
deployed on up to 100 endpoints.
3. Complete the
2 or more weeks
Broadly distribute the
XDR agent throughout the organization until all endpoints are protected.
4. Define corporate policy and protected processes.
Up to 1 week
Add protection rules for third-party or in-house applications and then test them.
5. Refine corporate policy and protected processes.
Up to 1 week
Deploy security policy rules to a small number of endpoints that use the applications frequently. Fine tune the policy as needed.
6. Finalize corporate policy and protected processes.
A few minutes
Deploy protection rules globally.

Recommended For You