Key Criteria for Evaluating Kubernetes Data Protection Solutionsv3.0

1. Summary

Kubernetes has become the de facto computing platform across on-premises, cloud, and edge environments, for stateful and stateless applications alike. As organizations move production applications to Kubernetes, platforms must support all kinds of application architectures, from monolithic three-tier applications to stateful databases.

However, organizations discovered that Kubernetes had critical limitations in day-2 operations around data storage and data protection. It turns out the Kubernetes project had underestimated the importance of persistent and reliable data storage and data protection. Although the problem was recognized a while ago, the development of solutions capable of properly addressing data storage and data protection for Kubernetes is still an ongoing challenge. In some cases, the solutions are immature, and it’s not enough to simply copy and paste virtualization data protection solutions to the Kubernetes world.

These difficulties were exacerbated by the evolution of multicloud strategies, which introduced additional challenges, and users are now asking for solutions that can work seamlessly across different providers and provide data migration and disaster recovery functionalities as well as additional security features to address threats such as ransomware attacks. Luckily, a new generation of maturing container-native solutions is kicking in.

We covered data storage for Kubernetes in our recent report, “GigaOm Radar for Data Storage for Kubernetes.” To get a full picture and build enterprise-grade, business-critical infrastructures and services, it is necessary to look at all aspects of data management, including backup and recovery, disaster recovery, and security. Kubernetes is still immature in some of these areas, even as users are becoming more aware of its shortcomings, so they are demanding a comprehensive data management strategy that can cope with large enterprises’ business, operational, security, and compliance requirements, such as those shown in Figure 1.

Figure 1. Cloud-Native Data Storage and Protection for Kubernetes Converge in Data Management Solutions

With this emerging demand in mind, solutions focused on data storage or data protection integrate or evolve to become data management solutions. Most of these solutions already abstract the storage access and protection layers from physical resources. They offer advanced data services beyond the basic functions of data storage or backup and recovery and provide a consistent user experience across multiple cloud platforms.

Traditional backup platforms are quickly adapting to offer solutions for Kubernetes as well. Their main advantage comes from the fact that they are already protecting the rest of the infrastructure and can be quickly adopted for Kubernetes clusters. However, most of these solutions were not originally designed to cope with the sheer number of variables and operations necessary to manage a complex Kubernetes environment; they lack the flexibility and efficiency to work with highly dynamic microservices-based applications.

Furthermore, most of these solutions were designed to work on-premises rather than in multicloud environments, which are becoming more common today. That said, traditional solutions are closing the gap, and most of the differentiation is now evident in the level of sophistication of data management features.

The GigaOm Key Criteria and Radar reports provide an overview of the Kubernetes Data Protection market, identify capabilities (table stakes, key criteria, and emerging technology) and evaluation metrics for selecting a Kubernetes Data Protection platform, and detail vendors and products that excel. These reports give prospective buyers an overview of the top vendors in this sector and help decision-makers evaluate solutions to decide where to invest.