Scalable, Cost-Effective Cloud Disaster Recovery with AWS DRS (Elastic Disaster Recovery)

Introduction

Ever since the evolution of information technology, there has always been a requirement for Disaster Recovery methods and techniques.

Various industries like mining, agriculture, education, and Information technology require a disaster recovery system for their purpose. Since the evolution of technology in the cyber world, various cyberattacks, internal security breaches, and natural calamities effect. It is necessary to ensure data integrity, back up the clients’ essential data, and protect our data and businesses from such an unfortunate crisis.

Since the pandemic curve has lowered, there has been a high rise in demand for information technology in business. Therefore, the need for a good disaster recovery strategy is a must to cope with any future unfortunate incidents.

Today we will be talking about configuring a service from Amazon known as AWS Elastic Disaster Recovery – DRS. This precious service helps to failback the AWS server resources and the on-premises server resources with the help of point-in-time recovery (RTO and RPO), minimized compute power, and affordable storage costs provided by the AWS.

Regions Supported

For now, these are the supported regions where you can do back up of your data through elastic disaster recovery:

How it Works

With minimal downtime, compute resources, and affordable storage costs, AWS Elastic Disaster Recovery service provides power to recover your Linux or Windows servers, either on-premises or AWS itself.

Architecture Diagram

For cloud disaster recovery, a replication agent is installed on the server to be backed up. All their volumes will be backup up as a point in time snapshot volumes in the region you configured for disaster recovery. A lightweight EC2 instance will be in always running state in AWS VPC, coordinating with the system volumes and replicating the data. Point in time snapshot volumes will be available based on the settings configured according to the mentioned retention period. You can recover the servers by a failback recovery approach which will use the volumes to create an EC2 instance of your choice of computing power. However, the storage used will always be equal or greater than the one in the server to be backed up.

With a new EC2 instance now launched up, you can use the EC2 to get your application up and to run or perform recovery back to the on-premises server with the help of a replication agent, which was installed initially.

Let’s follow the steps for AWS Elastic Disaster Recovery

1. Choose your On-premises or AWS EC2 machine, which has to be backed up
   Here we are using a ubuntu 18.04 machine for the demo
   The communication between source servers and Elastic DRS takes place over port 443 and TCP port 1500; make sure you have these ports open in the security group
   
2. SSH into the machine install the replication agent given by AWS elastic disaster recoveryWe will also make a demo text file which will be recovered at the end of the task
   
3. For Installing replication agent on the source serverRun the following command below and replace the region within which you want the replication to happenYou will also be required to be the root user for the processInstall the downloaded agentFor different operating systems, follow these steps: https://docs.aws.amazon.com/drs/latest/userguide/adding-servers.html/
   
4. AWS Elastic disaster recovery agent console
   You would need to set up the configuration for replication servers for further replication from the AWS DRS service
   Here you can select storage type, security group, data routing, etc. properties for your lightweight replication server
   
5. With all the correct replication server settings and replication agents installed in the source servers, you will find the progress of the replication status
   
6. You will now find the source server in Ready state for Recovery
   
   Also, you will find a lightweight instance of your choice as a replication server that will communicate with the source servers at all given times in your EC2 section
   
   Your replication server will be attached to 2 or more volumes at a given time. One will be the root storage, and all others will be a point in time captured volumes of your source servers
   
7. Recovery
   You can now perform a recovery or a drill to recover your source server in case of a disaster
   Choose the point in time you want to recover your source serverYou will see a conversion server that converts your data in captured volumes for use in an EC2 instance, this will be there in your AWS region for a few minutes and will terminate automatically as soon as its job is done
   
   After some time, you can see your backed up server launched in the AWS region, VPC, subnet you chose in the launch template associated with the AWS DRS service which can be modified at any given point in time
   
   You can finally SSH into the new machine and find your files backed up
   

Summary

The need for a disaster recovery plan for servers you are using is more important than ever but at the same time also affordable with AWS Elastic disaster recovery service. In addition, one can back up his data to other far regions, which will come in handy during the data center crisis. AWS DRS is quick, reliable, and affordable with minimal latency synchronization because of state-of-the-art storage devices and computing power. Not only do services like these help in disaster but also for moving data from one region to another or for any internal loss of data for both on-premises and AWS-backed servers.

About CloudThat