Veeam buys Kasten to get a boost in Kubernetes backup

Veeam has announced it is to acquire market-leading Kubernetes container backup specialist Kasten for $150m in cash and equity.

The move is logical for Veeam, which made its name as a provider of backup for virtualised environments and came from almost nowhere to rival the incumbent backup software providers.

Veeam’s chief technology officer (CTO), Danny Allan, said the Kasten acquisition would be “a key, pivotal milestone in Veeam becoming a $2bn company”.

He added: “People say the major shift has been from physical to virtual to cloud, but I think the real shift is from physical to virtual to containers. Containers provide a different infrastructure that has a lot of value and potential.”

Veeam will integrate Kasten’s K10 into its Veeam Backup & Replication and Cloud Data Management platforms.

Veeam and Kasten have been in partnership since earlier this year and, according to Allan, the early phases of technical and commercial integration of the two companies will see a continuation of processes that were started then.

Allan portrayed the likely integration path of Kasten as one where its backup functionality at the application/container end will be allied to Veeam’s capabilities in the back end.

This largely means a broader range of targets for backup data than was previously possible with Kasten.

Currently, Kasten uses object storage as a target for backups taken. Veeam is currently working on allowing that data to go to a Veeam repository. So, for example, Kasten will be able to send backup data to a data deduplication target, or tier via on-premise systems to the cloud.

Allan couldn’t say when that integration would be finished, but the capability has been demonstrated recently at VMware’s VMworld event.

Kasten will continue to be available as a standalone product for those that want to use it as such, however.

So how does Kasten work as a Kubernetes backup product? According to Allan, the key thing about Kasten is that it monitors Kubernetes storage locations from the application level. That is, from the level of the container, which means it can track and backup from numerous sources.

“Most data protection products in Kubernetes point to a file systems or a snapshot of a LUN,” said Allan. “With an application-centric view, you can have data in numerous environments.”

Other data protection product suppliers with Kubernetes support include Asigra, Cohesity, Druva, IBM Spectrum Protect, Trilio and Zerto.

Kasten’s Kubernetes protection is Agentless, as is that from Asigra, Trilio and Storware’s vProtect. Druva installs an agent on Kubernetes clusters, which leads to a resource overhead of about 1% to 2%.

Some analysts have said it can be difficult for data protection suppliers to develop Kubernetes support. Traditionally, backup is time-based and aims to restore environments to its state at a given point in time. Containers, however, are event-based and backing them up means capturing their state during event triggers.

Kasten was purpose-built for Kubernetes and provides a GUI management interface as well as an API that organisations can use to create custom offerings.

Containers are sweeping the datacentre, and provide a lightweight and rapidly developable and deployable means of running applications. They are essentially a form of virtualisation that provides all that is needed to rapidly start, scale up and scale down applications between on-premise and cloud environments.