Hammerspace today added data protection capabilities to its namesake data-as-a-service platform that makes it easier to manage data in environments running containerized applications.
Brendon Wolfe, vice president of product marketing and strategic alliances, says many DevOps teams want to manage the data protection themselves rather than rely on storage administrators, who protect data using legacy processes driven by a graphical user interface (GUI). Containerized application environments are too dynamic for storage administrators to protect using those legacy tools, he notes.
Hammerspace employs machine learning algorithms to optimize data management across hybrid cloud computing environments, including IT environments that have adopted Kubernetes. The latest edition of Hammerspace adds the ability to globally undelete files and snapshots so DevOps teams can self-service their own data recovery requirements.
In addition, Hammerspace has added support for automated data classification, integration with customer-managed key management systems and the ability to harvest metadata from cloud analytics services to detect and tag files.
Wolfe notes as more stateful applications built using containers that need to access persistent storage find their way into production environments, data protection issues will become a more pressing issue. Right now, developers and storage administrators are expecting the other to solve this issue. By providing a separate control plane away from the underlying data, Hammerspace is making it possible for DevOps teams to manage data protection without having to wait for an IT operations team to figure out how to apply legacy policies to this new application paradigm, he says.
Arguably, the biggest challenge with data protection in containerized applications is the rate at which microservices based on containers are updated. In some cases, containers might be ripped and replaced several times before a scheduled legacy backup process kicks off. That means, in theory, containerized applications can be non-compliant multiple times a day. The Hammerspace data-as-a-service platform makes it easier for DevOps teams to back up data whenever they deem appropriate, Wolfe says.
Many organizations are underestimating the amount of file data they will need to manage in containerized application environments. As such, it’s only a matter of time before DevOps teams and the IT staff in charge of DataOps find themselves at odds with one another. Hammerspace is making a case for a separate data management platform that resolves data management conflicts long before they get a chance to come to a head.
It may be a while before DevOps teams building and deploying containerized applications find themselves in meetings with DataOps teams to work out these issues. In the meantime, however, many DevOps teams may just go ahead on their own. After all, most organizations are a lot more concerned with whether data was protected than they are with how it was done.