Best Practices for Disaster Recovery in the Cloud

30 July 2020 | Posted by Matt Willis

From environmental upheavals like hurricanes to terrorism and cybercrime, massive data loss is a threat to any global enterprise. Yet it's not only data that's lost when a system goes down - revenue is affected too. Companies like Amazon lose tens of thousands of dollars a minute during downtime. Not only that, but downtime hurts the reputation of a business, causing it to look unreliable. Effective cloud disaster recovery protocols get you back up and running quickly. 

Being able to control how you respond to disaster dictates how successful your organization will be at recovery, so when disaster does happen, your lost data is resorted to the local data center or cloud provider instead.

Cloud disaster recovery

The goal of disaster recovery is to protect business assets and it's key to a business continuity strategy. Cloud providers have created dedicated facilities to offer effective backup capabilities, giving businesses access to scalable, autonomous disaster recovery services.

When it comes to creating a cloud disaster recovery strategy, there are two things to consider - your recovery time objective (RTO), and your recovery point objective (RPO). RTO is the amount of time it takes your business to become fully operational again after a disaster, while RPO is the amount of data your organization is willing and prepared to lose in the disaster recovery process.

For example, if an organization has a cloud infrastructure taking automated backups every 5 minutes in one region, and then a warm standby of the current infrastructure in a second region that takes 20 minutes to come online, your current RPO is 5 minutes, and your RTO would be 20. The lower a business wants these numbers, the more costly the solution.

Learn how effective cloud management services can help you save costs and  improve security in the cloud. Download our brochure today.

When do disasters happen in the cloud?

If you’ve already migrated your assets to the cloud, or are in the midst of a migration, it’s crucial to understand how disasters can happen to prepare yourself. These include:

  • Malware – Hackers are able to generate threats to your online infrastructure, and they’re evolving all the time.
  • Ransomware - Having a strong backup strategy can protect potential losses against a ransomware attack. 
  • Equipment failure – If your technical infrastructure were to fail, do you have a contingency?
  • Human error – People make mistakes, and even simple errors can lead to major issues that affect multiple end-users.
  • Natural disasters – You need your organization to be prepared in the event of a natural disaster. You can't control it, but you can limit its impact.

How to prepare for disaster

The best way to prepare for a massive loss of data is to regularly test your disaster recovery scenarios. 

There are a number of potential backup tests that you can run within your organization:

  • Walkthrough test – A basic review of your plan, the walkthrough test ensures everyone involved remains updated.
  • Parallel test – Using a virtual machine to ‘restore’ your system, you can observe how effective the recovery is.
  • Tabletop test – Members of different departments explain what to do in individual disaster scenarios, revealing potential shortcomings in the existing strategy.
  • Full interruption testing – The most in-depth test, full interruption testing risks actual downtime as it exposes how your business would actually fare after unexpected data loss.

If your business relies on cloud backup, Data Recovery as a Service (DRaaS) offers the usual powerful backup and recovery services, but also restores your cloud-based infrastructure, application, and storage to a ‘hot’ site while your IT department rebuilds your infrastructure after a loss. This reduces your exposure to risk, ensuring your mission-critical computing resources are back online as fast as possible.

Benefits of operating disaster recovery in a multi-cloud environment

Having your resources attached to two different providers means there’s virtually no chance they’ll both experience the same disaster – meaning you’ll always have your resources available somewhere.

When operating in a multi-cloud environment, public cloud providers offer more reliability than local data centers but could still suffer in the wake of a major disaster. Multi-cloud disaster recovery allows you to duplicate your resources to a second cloud provider in another geographic region. 

Best practice for disaster recovery

Best practice for disaster recovery encompasses both company policy and technical testing. Your disaster recovery plan should include a risk assessment and a business impact analysis, and you should choose and implement prevention and recovery measures. 

When creating a disaster recovery plan, you should adhere to the best practice in order to safeguard your data, while also understanding your RTO and RPO objectives, which should be based on your business needs:

  • Perform your due diligence 
  • Choose a disaster recovery planning method
    • Backup and restore allows you to use a managed solution to backup and restore your data as and when you need to. Backup and restore has an RPO of minutes and an RTO of hours.
    • Pilot Light focuses on a core subset of critical applications and data, keeping them running to ensure they can be quickly retrieved. Pilot Light has an RPO of minutes and an RTO of hours.
    • Warm standby duplicates the system’s core elements and keeps them running on standby concurrently with the rest of your system. Warm standby has an RPO of seconds and an RTO of minutes.
    • Hot standby creates a full replica of your data and applications and deploys it in two or more active locations. Hot standby has an RPO of seconds, or possibly none at all, and an RTO of seconds. 
    • One key thing to note is the importance of storing backups in separate regions or availability zones. Your strategy will be rendered useless if the region your data is stored in goes down. 
  • Keep your disaster recovery plan up to date to anticipate evolving threats.
  • One benefit of working in the cloud is the ability to automate almost anything. Automation in a disaster recovery strategy allows for scheduled configuration backups, prepared for any situation. It can also be used to automatically copy backups to other regions.
  • You can also use some of the features native to your cloud provider's storage as part of your disaster recovery strategy. For example, the Simple Storage Service (S3) offered by Amazon Web Services stores your data by default across 3 availability zones (different data centers in the same region). However, remember that if an entire region goes down, you won't be able to access your data. For this scenario, S3 Cross Region Replication can be used to automatically copy your data to a different AWS region. S3 also has a versioning feature that helps protect against accidental deletion or even ransomware attacks. Just remember that with both versioning and cross region replication, you'll be charged for the additional storage used.
  • Test and update your disaster recovery plan on a regular basis. 
  • Maintain your disaster recovery plan to keep up with system changes that might affect you.
  • Select the right cloud support provider for your business.

What to look for in a cloud disaster recovery provider

When selecting a cloud disaster recovery provider, your business needs to consider several factors before deciding.

  • Physical location. Having your disaster recovery provider located too far away from your organization puts you at risk of latency and congestion, making it difficult to access your restored content.
  • Reliability. Your cloud disaster recovery provider must be reliable, as downtime during the recovery process may lead to disaster.
  • Scalability. A cloud disaster recovery provider needs to be able to not only protect selected data and applications, but also accommodate additional resources as needed and provide sufficient performance even as other global customers use the service.

How Cass can help with disaster recovery

With disaster recovery itself moving to the cloud, you can minimize downtime and protect your files by leveraging a recovery solution. We’re able to provide support to organizations that have recently migrated their processes to the cloud or are in the middle of migration.

Our agnostic cloud expertise means we can help your business create the ideal disaster recovery plan and point you towards the cloud recovery provider that will meet your requirements.

New call-to-action

Topics: Cloud Management Services

Get weekly Cloud roundups direct to your inbox.