A Question for Kubernetes: Enabling Data Storage
In a rapidly changing and unpredictable enterprise technology landscape, flexibility rules the day. That’s why businesses of all sizes are embracing hybrid cloud strategies for data and application mobility and agility. Enter Kubernetes, a portable, extensible, open-source platform for managing containerized workloads and services. In the latest GigaOm research report, Key Criteria for Evaluating Kubernetes Data Storage, analysts Enrico Signoretti, Max Mortillaro, Arjan Timmerman examine the ongoing storage questions around Kubernetes adoption.
Kubernetes has been gaining popularity for several years, but the COVID-19 pandemic in 2020 put that movement into overdrive, forcing many organizations to come up with contingency plans to support their transformed activities, business processes, and users. The agility, flexibility, and efficiency offered by the Kubernetes platform meant that demand for scalable and enterprise-ready Kubernetes storage solutions spiked.
“Every organization accelerated their digital transformation initiatives during the pandemic, and moved to the cloud for added flexibility. New applications are now developed with containers and microservices in mind and Kubernetes is the platform to run them,” says Signoretti. “The reason is as simple as that: Kubernetes was just the right technology at the right time and many developers were already evaluating whether to adopt it. COVID simply helped them to make the final decision.”
As a result, Signoretti says Kubernetes, and containers and microservices in general, are becoming mainstream. Organizations have moved past the evaluation phase and are now performing early production Kubernetes deployments. But reliable data storage, as well as data management and security, still remain key factors to consider, he says. The interface between the containers managed by Kubernetes and the underlying storage infrastructure is enabled through Container Storage Interface (CSI). Signoretti says Kubernetes needs the right integration with infrastructure layers, and IT leaders should take note in assessments and figure out what they may need for storage.
“When it comes to storage, Kubernetes and CSI offer low-level integration that is good for basic operations but lack the sophistication necessary to support complex scenarios. It is up to the storage vendors to offer additional features and integrations to build the right foundation for data mobility and enterprise-grade data services,” he says.
As Kubernetes adoption continues, enterprises are willing to onboard more critical applications in their container initiatives. They are demanding better consolidation opportunities and enterprise-grade features such as advanced data protection, replication, better mobility, and disaster recovery.
When performing early production Kubernetes deployments, Signoretti advises starting slow.
“Most organizations take a conservative approach,” he says. “They choose applications that are not business critical first and start with those. This helps to build the necessary expertise without too much pressure. When the first applications scale up and put some pressure on the infrastructure, the user starts to understand better the complexities, challenges, and potential solutions.”