The AppDynamics Controller is certified to run on an AWS environment with the Aurora Database. This page provides information on installing, configuring, and administering a Controller deployment on AWS with Aurora Database. 

Installation Overview

You can deploy a medium- or large-scale Controller in AWS using Aurora. Aurora provides higher performance than MySQL, which allows you to scale your Controller to handle more metrics.

Before you install the Controller, review the requirements for the components you plan to install and prepare the host machines. The requirements vary based on the components you deploy and the size of your deployment.

You can manually deploy your Controller on AWS. See Deploy the Controller on AWS 
This deployment requires attention and time because you have to set up all of the configurations. However, this gives you freedom to customize your deployment.

You use the Enterprise Console to deploy the Controller by specifying Aurora as the database type. The Enterprise Console is the installer for the Controller, and you can use it to manage the entire lifecycle of new or existing AppDynamics Platforms and components.

You can get the software for installing the platform components from the AppDynamics download site. See Download AppDynamics Software

AppDynamics on AWS Architecture

The following diagram depicts the components of an AppDynamics Controller deployment on AWS.

AppDynamics AWS Architecture

Configure an Application Load Balancer in front of the Controller and ensure SSL terminates at the Elastic Load Balancer. 

With Amazon Relational Database Service (RDS), you no longer need to worry about database backups, as it takes care of this for you. Also, you no longer need to implement high availability (HA) on your own, since you can instead leverage the Standby Replica that Aurora/RDS offers and the Aurora database is horizontally scalable. With the multi-AZ deployment option, Aurora offers 99.95% availability.

Controller High Availability and AWS

    • If your data is migrated to the Aurora DB, you can create a new Controller from the Amazon Machine Image (AMI) in case of failure. With Aurora as the database, HA scenarios are not required because the Aurora database is horizontally scalable. 

    • It is a good practice to cut the new root AMI every time you make a configuration change to the Controller.

    • You can configure a read replica instance while creating Aurora DB, which should satisfy most database replication requirements.

    • You can configure the backup policy while creating the Aurora DB and modify it later.