Chapter 2. New features

OpenShift Data Foundation (ODF) on a two‑node OpenShift cluster with fencing enabled is now supported. This architecture is suitable for distributed and edge deployments, where the goal is to minimize hardware footprint while maintaining High Availability (HA).

For more information, see Deploying OpenShift Data Foundation on two-node clusters.

Erasure coding support for OpenShift Data Foundation (ODF) internal mode helps to enhance storage efficiency and reduce infrastructure costs. Erasure coding enables more space-efficient data protection compared to traditional replication, making it an ideal solution for customers seeking scalable and cost-effective storage.

For RADOS Gateway RGW object storage, erasure coding can only be enabled when creating the OpenShift Data Foundation cluster.

For RBD and CephFS storage, erasure coding can be enabled during cluster creation or later by creating a custom StorageClass.

For instructions on enabling erasure coding, refer to the deployment guide applicable to your environment.

S3 Vectors provide native vector-database storage capabilities within the Multicloud Object Gateway (MCG). Vector buckets enable you to store and manage vector embeddings for machine learning and AI applications, backed by NSFS (NamespaceStore FileSystem) storage.

For more information, see S3 Vectors in Multicloud Object Gateway.