By Chris Williams, GreenPages Enterprise Cloud Consultant
First Look at AWS Backup
One of the glaring misses in the AWS portfolio had been a consolidated, concise mechanism for backing up all yo’ stuffs without having to go 3rd party. While I am a *huge* fan of the Veeam product line… as of now you can’t use it to back up AWS RDS (as well as several other native services), so creating a backup strategy for your cloud services has generally been a mixed bag of concessions, stopgaps, and multi-product/process approaches if you are using more than just EC2 and S3. Sure you can use the RDS native backup options, but what about when you want to have a crash consistent (or even better application consistent!) backup of your whole stack from a certain point in time? Queue AWS Backup. In this article I’m going to talk about what it is and isn’t.
What it is:
Straight from thewebsite: “AWS Backup is a fully managed backup service that makes it easy to centralize and automate the backup of data across AWS services in the cloud and on premises.”
Translation: AWS Backup is a one stop shop to run backups for the following services:EFS, DynamoDB, EBS, RDS(except Aurora) andAWS Storage Gateway. By using resource tags (nice!) you can specify backup windows, retention periods, and data life cycle policies. AWS says that they will also backup on-prem data, but the caveat here is that it needs to be on the storage gateway.
Effectively what AWS Backup does is manage the pre-existing backup mechanisms for the above services. If you were already running RDS backups & wrote a script to trigger EBS snaps… then this will give you a one-stop-shop to manage said backups/restores.
One really nice things about AWS Backup is the EFS backup feature. Previously if you wanted to backup your EFS environment using native AWS services, you had to do anEFS-to-EFS backup solutionthat…. wasn’t great. It’s a CFT that you deployed into your environment and subsequently configured. It *works*… but it’s a manual process & just another thing you need to keep an eye on. By incorporating all of the backup service offerings into one service, AWS has made it much easier to get point-in-time, crash consistent backups of the above services.
What it isn’t:
Because this isn’t a new technology under the covers, the limitations of the backup offerings of the above AWS services carry over into this new management console:
Backups are crash consistent, not application level consistent.
If you have other services other than those listed above, you’ll need to use a different backup strategy.
It’s not DR: While it is a good tool in an overarching BC/DR strategy, it won’t get you there on it’s own. If you want to go with a multi-region strategy you’ll need to look elsewhere.
AWS backup is a great 1st step down the road of helping customers flesh out a better native BC/DR solution, but it is exactly that, a 1st step. There are a lot of features and functionalities that I want from a more robust backup solution. I expect that there is a lot of back and forth within AWS about how robust they can/should make this offering before they start stepping on the toes of their backup partners.
AWS Backup is an excellent option for someone who is:
running only the above services
in a “less than highly complex” environment
doesn’t need application level consistency
doesn’t need multi-region DR
Strategy, architecture, integration & management across the world’s leading technologies.