Hello Team,
We are exploring solutions for node-level data replication when using SPDK (Storage Performance Development Kit) as the backend for LINSTOR, particularly in an OpenStack environment.
As the LINSTOR documentation section below highlights, when leveraging the NVMe-oF/NVMe-TCP LINSTOR layer, the resources are not replicated because LINSTOR is not using DRBD in this configuration. The data resides on a single node.
3.5.1. NVMe-oF/NVMe-TCP LINSTOR Layer
NVMe-oF/NVMe-TCP allows LINSTOR to connect diskless resources to a node with the same resource where the data is stored over NVMe fabrics. This leads to the advantage that resources can be mounted without using local storage by accessing the data over the network. LINSTOR is not using DRBD in this case, and therefore NVMe resources provisioned by LINSTOR are not replicated, the data is stored on one node.
My understanding is that a native, integrated LINSTOR-based replication solution (like DRBD) does not currently exist for resources backed by SPDK via NVMe-oF/NVMe-TCP.
Question: Is there an existing or alternative solution for achieving synchronous or asynchronous data replication at the node level for LINSTOR resources provisioned with an SPDK backend?
We need to address this replication gap to ensure high availability and data durability for our OpenStack volumes.