Other hyperconverged products commonly consume individual drives using what are typically referred to as disk groups.

This generally consists of multiple HDDs (typically up to six) paired with a dedicated SSD that functions as a cache disk. If one drive in the group fails, the entire disk group is normally marked as failed.

Many of these solutions recommend a minimum three-node deployment because a two-node deployment often provides only node resiliency with no internal drive resiliency. In other words, a single drive failure can result in the entire node becoming unavailable.

SvHCI and SvSAN are designed differently. Since they are targeted toward highly available two-node deployments, they consume storage using a traditional RAID model, providing internal disk resiliency through either hardware RAID or software RAID while also maintaining cross-node resiliency through synchronous mirroring (Replication Factor 2).

This enables both SvHCI and SvSAN to tolerate the failure of an individual SSD or HDD without immediately losing the storage pool.

Both products also implement a Storage Failed recovery mechanism.

Video Resource

The following video demonstrates a complete end-to-end deployment.

StorMagic SvHCI supports software RAID when a hardware RAID controller is unavailable.

From the SvHCI interface, navigate to Storage > Pools.

In the example below, two 3.94 TB Samsung M.2 NVMe SSDs are configured as a software RAID 1 storage pool using SvHCI.

SvHCI can also consume logical disks presented by a hardware RAID controller. The RAID controller manages the physical disks while SvHCI provisions virtual machines on the presented logical volume.

StorMagic Product Documentation