Author: Dave Russell, VP, Enterprise Strategy, Veeam
Unbeknown to most, containers have been around since the 1970s. But it wasn’t until the advent of Docker in 2013 and Kubernetes in 2014 that their popularity began to explode, quickly becoming the go-to packages of software to build, test, deploy and redeploy applications on multiple environments.
Together, these platforms have revolutionised the way companies use containers – unlocking a whole world of benefits in the process, with their ability to reduce overheads, while increasing portability, consistency and efficiency – and ultimately driving the containerisation market’s expansion and movement toward maturation.
Such is this expansion, 27% of organisations are already using containers, with a further 34% in the planning and testing phase of container deployment, according to Veeam’s 2021 Cloud Protection Trends Report. Only reinforced further by the application container market’s expectation to register a CAGR of 29% over the forecast period 2021 to 2026.
In fact, only 16% of organisations currently have no interest in deploying containers at this time. But while more and more businesses are beginning to realise the potential of containers and grow excited about the benefits they can bring, there’s an essential, mutually inclusive component that cannot be overlooked: data protection.
This is where businesses looking to reap the benefits of Kubernetes and containers require education – and herein lies the purpose of this article. Below, I explore the pressing need for data protection to keep pace with the maturation of containers, and how businesses can accelerate their modern data protection strategy.
Understanding the container data protection landscape
In today’s increasingly digital world, as computing and storage rapidly move to the cloud, containers will likely become a vital technology for every modern organisation. However, while container orchestration tools, such as Kubernetes, are convenient for their scalability and portability, they can fall short when it comes to data protection.
Reasons for this vary. But a major challenge lies in the fact that because containers by their very nature are designed to be temporary, organisations are realising their lifespans are often shorter than the data they create. So, for reasons of compliance, cybersecurity, or protection against system outages, data must be backed up and stored.
Similarly, replication is required for migrating entire clusters between testing, development and protection environments, to be able to return to previous states quickly if you don’t get the result you’re hoping for. But across the board, more organisations are discovering unexpected things can happen during migration, only strengthening the case for data protection and backup.
It’s therefore promising to see that of those organisations deploying containers, 46% back up in some capacity, while 8% are actively seeking a solution, according to the Cloud Protection Trends Report. Perhaps highlighting how organisations understand that Kubernetes is not immune to data protection failures.
For the minority who for whatever reason do not back up, however, 26% believe their container architecture is natively durable and 17% don’t believe they store stateful data. Unfortunately, both of these statements are untrue – regardless of platform, data loss scenarios still take place in Kubernetes which are not addressed by storage availability or replication.
In response, Kubernetes environments require an application-centric approach over an infrastructure focus – and organisations need a backup solution that works against a wide range of Kubernetes application stacks and deployment methods. Establishing a container data protection strategy and following backup best practices can be the key to realising this.
Five best practices for backups
In the wake of container maturation, many organisations are proving they don’t understand how containers are uniquely designed. Of course, needs and approaches to overcoming this will vary from business to business. But underpinning any effective container data protection strategy is backup solutions, centred on: architecture, recoverability, operations, security and portability.
Architecture: the platform used to protect Kubernetes applications needs to automatically discover all the application components running on your cluster and treat the application as the unit of atomicity. In addition, it must include the state that spans across storage volumes and databases, as well as configuration data included in objects, such as ConfigMaps and Secrets.
Recoverability: data management platforms must allow you to restore the application components you want, where you want them. Granularity is also key to restore only an application subset, such as the data volume. But above all, the approach must make restoring simple and powerful by also allowing you to select the appropriate point of time copy of the application.
Operations: it’s imperative a Kubernetes-native backup platform can be used at scale and provide operations teams with the necessary workflow capabilities while meeting compliance and monitoring requirements. Operators should be able to offer self-service capabilities to application developers without requiring application code or deployment changes.
Security: controls around identity and access management and role-based access control (RBAC) must be implemented. RBAC permits different personas in an operations team to adopt a least-privilege approach to common tasks, such as monitoring. Encryption at rest and in transit must always be employed to ensure data is secure whenever it has left the compute environment.
Portability: in a multi-hybrid-cloud world, a cloud-native data management platform needs to be flexible in the support of multiple distributions, and offer capabilities that allow for the portability of workloads and applications across these diverse environments. Portability capabilities are required across multiple use cases, including application restoring, cloning and migration.
Accelerate your container data protection strategy
By following the above best practices, you’ll be well on your way to meeting the backup, restore and mobility requirements of your entire Kubernetes application. But as containers continue to mature, partnering with third-party experts will be the most effective way to address the cloud-native modern data protection needs for enterprises.
