Disaster recovery for Kubernetes on the HPE Alletra replication you already run. HPE Remote Copy keeps a copy of your volumes at the DR site. At failover, CloudCasa brings your Kubernetes workloads and KubeVirt VMs up on the DR cluster from those replicated volumes, and moves them back when the primary site is ready. Nothing is restored from backup.
Talk to a DR ExpertCloudCasa DR works with infrastructure you already have in place: HPE Alletra arrays replicating with HPE Remote Copy, and a Kubernetes cluster at each site. There is no separate replication layer to add, and nothing is rebuilt from backup. CloudCasa records what runs on the production cluster and, at failover, brings those workloads back up on the DR cluster using the replicated volumes.
For teams running Red Hat OpenShift, OpenShift Virtualization, SUSE Rancher, or other Kubernetes platforms supported by the HPE CSI driver, CloudCasa adds the Kubernetes side of disaster recovery. Existing storage replication becomes repeatable failover and failback for both containers and KubeVirt virtual machines.
Each site has a Kubernetes cluster and an HPE Alletra array. HPE Remote Copy keeps the data in sync between the arrays. When you fail over, CloudCasa uses the replicated data to rebuild your workloads on the DR cluster.
CloudCasa DR for Kubernetes, powered by HPE Alletra Storage MP B10000, combines replication on the arrays with Kubernetes-aware failover and failback.
HPE Alletra replicates the data, so it is already at the recovery site. Failover does not restore from backup or copy data between sites.
Protect Kubernetes applications and KubeVirt virtual machines in the same DR Plan, with the same failover and failback process.
Failover uses volumes that are already replicated, so recovery typically takes seconds to minutes rather than hours. Actual time depends on the arrays and how quickly the workloads start.
Failover and failback run as DR Recovery Jobs from a DR Plan you define in advance. By default, CloudCasa stops the source workloads first, then handles the Remote Copy group operations on the arrays.
RPO is set by the Remote Copy mode and schedule on the arrays, not by a backup interval.
CloudCasa rebuilds the Kubernetes resources, PVC bindings, and KubeVirt VMs from the replicated inventory, so workloads come back online without manual rebuilding. Each recovery covers every volume in the affected Remote Copy groups.
Built for platform engineering and infrastructure teams running critical Kubernetes workloads.
Four steps from replicated storage to a working DR site.
Set up HPE Remote Copy between the production and DR HPE Alletra arrays.
Register the production and DR clusters with CloudCasa DR enabled, add both arrays as storage systems, and validate them.
Build DR Plans that protect whole namespaces or selected Deployments, StatefulSets, and VMs.
Run a DR Recovery Job. CloudCasa fails over the Remote Copy groups and recreates the Kubernetes resources and VMs on the target cluster.
CloudCasa stops the source workloads by default, fails over the Remote Copy groups on the arrays, imports the replicated volumes into the DR cluster through the HPE CSI driver, and restores the Kubernetes resources, PVCs, and VMs there.
CloudCasa runs the Remote Copy recover and restore steps, waits for resynchronization when needed, returns the arrays to their original roles, and brings the workloads back on the primary cluster once the primary site is ready.
CloudCasa DR with HPE Alletra works on CSI-conformant Kubernetes distributions where the HPE CSI Driver for Kubernetes is supported, including Red Hat OpenShift, SUSE Rancher, SUSE Virtualization, and upstream Kubernetes. The HPE CSI driver must be installed on both the production and DR clusters.
Designed for multi-cluster operations, the solution helps standardize DR across different Kubernetes platforms.
No. Failover uses volumes that HPE Alletra has already replicated to the DR site. CloudCasa rebuilds the Kubernetes resources around those volumes. It does not restore data from backup.
Yes. Both run as DR Recovery Jobs. For failback, CloudCasa runs the Remote Copy recover and restore steps, waits for resynchronization when needed, and returns the workloads to the primary cluster.
Yes. A DR Plan can include Deployments, StatefulSets, and KubeVirt VMs, such as those running on Red Hat OpenShift Virtualization or SUSE Virtualization.
No. CloudCasa uses your existing HPE Remote Copy groups. Failover covers whole Remote Copy groups, so workloads that need separate recovery decisions should use separate groups. Confirm the Remote Copy mode you plan to use with CloudCasa Support before production.
No. Your storage team keeps managing the arrays with HPE tools. During failover and failback, CloudCasa calls the array API to run the Remote Copy group operations.
No. DR recovers workloads from the latest replicated state. Backup keeps point-in-time copies for longer retention and recovery from deletion or corruption. CloudCasa DR is available on its own or as an add-on to a CloudCasa backup subscription.
For the workloads in a DR Plan, CloudCasa recovers the Kubernetes resources, PVCs, PVs, required custom resources, and KubeVirt VMs on the target cluster. Changing IP addresses during recovery (Re-IP) is on the roadmap, so DNS and network changes outside the cluster remain your responsibility.
Protect critical Kubernetes applications with low RPO and RTO, using orchestrated failover and failback on HPE Alletra.
