Users of cloud-native storage in public clouds know that each processing cycle consumed by the storage application costs money. Thus, greater efficiency equals greater cost savings. Ondat, an enterprise-grade, container-native storage platform, has come out on top in a benchmarking study of the five leading storage providers for cloud-native applications.
In its report, “Performance Benchmarking Database Platforms and Cloud Native Storage Solutions for Kubernetes,” the research firm Architecting IT measured three characteristics: latency, performance (throughput) and consistency (deterministic performance). The first two metrics describe the efficient use of storage resources. The third looks at the variability of performance, highlighting the ability of each storage solution to provide a consistent level of I/O.
Five platforms were tested: Ondat, Portworx from Pure Storage, OpenEBS/cStor, Longhorn and Rook/Ceph. Ondat achieved the best results across all the tests, followed by Portworx. The study looked at three common structured data solutions (Postgres, Redis and MongoDB) and once again ran a series of tests to compare container-native solutions side by side. When testing on Postgres, the data showed Ondat has 32% faster database transaction throughput than Portworx from Pure Storage. Overall, commercial offerings in the test cohort significantly outperformed their open source alternatives.
Cloud-native applications orchestrated with Kubernetes need persistent storage managed from the cluster and applications directly. Container-native storage has emerged as a new category of software-defined storage where the traditional features of data storage (resiliency, data protection, scalability) are built into containers running inside a Kubernetes cluster. This enables deployments to use local storage resources, whether that be physical storage in an on-premises solution or native storage (including NVMe) in public clouds.