michelangelus - Fotolia
IT teams need a solid backup plan in place so they're prepared to react when a data storage component fails or data gets lost or corrupted.
That's where a service like AWS Backup can help. Added in January 2019, the AWS Backup service provides a centralized way to configure, manage and restore backups for multiple Amazon cloud services. It currently supports backups for Relational Database Service (RDS), Elastic Block Store (EBS), Elastic File System (EFS) and DynamoDB. It also supports AWS Storage Gateway volumes, so organizations can use it for data stored on premises as well.
Get started with the AWS Backup service
AWS Backup is relatively simple to set up and has an intuitive console experience, API and command-line interface (CLI). To start, create a Backup Vault -- a container that groups data backups -- or select an existing one. Then, create or choose a backup plan to set schedules, rules and determine which resources are backed up -- RDS databases, EFS file systems, Storage Gateway volumes, etc.
The service also supports lifecycle rules that define how long it will retain backups and when to move backups to cold storage. The cold storage feature is currently only available for EFS, but I expect it will expand to other services; cold storage is up to 80% cheaper than other storage types.
AWS users can define backup plans with JSON templates, which can be integrated into automation processes and code-versioned. Once a backup plan is defined, users keep track of backup activity by looking at the Jobs section in the AWS Management Console or using the AWS API or CLI. They can also trigger ad hoc backup jobs as needed and launch AWS resources from existing backups. The AWS Backup service supports Key Management Service encryption for all backups.
AWS Backup gives users storage flexibility and visibility
AWS users had several ways to manage data backups in the past, such as RDS snapshots, EBS snapshots and DynamoDB backups. But prior to AWS Backup, users had to configure these options individually. In other cases, such as EFS, there was no built-in mechanism, so users had to create and manage their own custom processes with Lambda, Data Pipeline or scripts.
AWS Backup simplifies those operational tasks and makes it easier to work with applications that have multiple data components spread across many Amazon cloud services. Once lifecycle policies are in place, resources can be tagged so they follow the appropriate backup plan.
It also provides more flexibility to configure lifecycle rules. For example, RDS supports a maximum retention period of 35 days for automated backups, while AWS Backup can keep snapshots for years or even indefinitely.
Pricing is the same whether backups are configured individually or through AWS Backup. The same applies to the cost of restoring data.
AWS Backup still needs deeper support
Keep in mind that AWS Backup currently only supports the subset of AWS data storage services mentioned above -- RDS, EBS, EFS and DynamoDB.
At this time, the AWS Backup service lacks support for a number of Amazon databases, such as S3, Elasticsearch, ElastiCache, Neptune, Redshift and DocumentDB. It doesn't support copying backups across AWS regions or accounts, which is something many disaster recovery implementations require.
There's also room for better infrastructure automation processes. Even though you can create backup plans with JSON templates, you can't yet manage AWS Backup resources with CloudFormation. CloudFormation support would bring consistency to existing infrastructure automation processes and make it simpler for users to add AWS Backup components to their existing templates or to new ones.
CloudWatch metrics are another missing piece. Currently, AWS Backup only supports Simple Notification Service alerts when jobs start and complete. This capability would deliver better visibility into backup activities.
Since AWS Backup is relatively new, I would expect AWS to announce additional features in the near future. However, even in its current form, this is a service that both saves users a significant amount of time and simplifies the critical operational task of data backup management.