At AWS Summit in Washington DC I had the honor of speaking about AWS, Data Protection, and Cohesity. When delivering this presentation, it was important to me to be able to present the real-world challenges from the shoes of the IT enterprise administrator. Let’s take a look!
Data Protection Challenges from the Real World
In this section I remain cloud-vendor agnostic, so you will not know which cloud offering they happened in. They did happen though, and it’s important to articulate that even though your data is in the cloud, it is your data, you are responsible for it.
- Cloud Instance Corruption
The Situation: There is nothing worse than doing your due diligence to apply windows updates, and then the system becomes corrupted and doesn’t come back on online. Even worse, then finding out the system owner didn’t know that they needed to enable backups manually on their own. For a production system the response can be panic. In this case support was able to recover the server, but it took over 24 hours. Yes, a production system was down for over a day due to lack of backups.
Lessons Learned: Thankfully the system was restored, but it’s important to consider a couple critical elements to the learning opportunities. Having a process to ensure a cloud-based instance is backed up can go a long way for the availability of your system and its data during an unexpected situation. Taking that a step further, knowing your Recovery Point Objectives (RPOs) and Recovery Time Objectives (RTOs) is important. Then taking that a step further is deploying an automated backup and recovery process to ensure recoverability of enterprise data.
- Ransomware
The Situation: A remote user was using the internet through an externally secured platform to do their job. While doing this work, they crossed paths with something that introduced ransomware to the system. In this case it was a cloud-based server providing the users application access. Thankfully the user acted quickly when they noticed something wasn’t right and reported it to the helpdesk. The server admin restricted access to the network immediately. While it wasn’t yet clear that it was ransomware, this step was important to prevent the spread of whatever was on the server. Due to both timely and good backups the server was restored, and the system admins were able to spend the time they needed identifying the ransomware.
Lessons Learned: Having good server backups in the cloud allowed for quick recovery and quick response. Also, kudos to the employee that was security educated and knew to involve the helpdesk quickly.
How AWS does data protection natively and key considerations
When backing up AWS EC2 instances there are a couple different ways to approach this without purchasing an official solution. Instance by instance or via the Lifecycle Manager
Backing up instance by instance – When using this option, the administrator is setting up the backup on a per instance basis and can be scheduled in the AWS snapshots console. Things to consider when using this option are that when looking to do a recovery, having proper tagging in place up front will make you more successful. There is a chance that someone will not take the step to setup a backup, which creates risk. This is yet another console that would be used to manage backups instead of just one.
Lifecycle Manager – This option is wizard driven. When proper tagging is used, multiple servers can be scheduled to back up in a job. Key considerations with this option are that if the proper tagging isn’t done up front, then some server backups could be missed. Similar to the above this introduces another console.
How Cohesity Simplifies Data Management
We also discussed how Cohesity for backups will allow you a single pane of glass for backup management of all your systems regardless of location (cloud, on-premises, etc). Our deduplication and compression abilities – available even within the cloud – are also a huge consideration. Our search/index ensures RPOs and RTOs are being met.
Over and above backups we also support archiving, test/dev, analytics/search, splunk and more. We also announced that we also work in GovCloud!
Final thoughts!
As you move into a journey with your data in the cloud it’s important to understand that your data is your data. As shared above, the unexpected can happen even in the cloud. Choosing a solution that will allow you to ensure that your data can be recovered and brought back online in an efficient way is imperative to long-term business success.
