VirtuRAID controller lends itself well to use cases that require processing large amounts of data, especially where the workload exhibits a small block random IO pattern. This is typically in a virtualized environment in the fields of genetics, video surveillance, analytics, oil & gas, GIS, and media.
There are two use cases for this card.
Server based PetaByte scale, high performance primary storage: A single VIRTURAID controller can RAID PetaBytes of SSDs and cache to Terabytes of host RAM. This raw storage can be used for the purpose of creating a Petabyte scale high performance repository on a single server, to be used as local storage in Linux, Linux based hypervisors and containers, shared storage for hyperconverged software, and (soon) datastores in VMware vSphere.
In-server TeraByte scale cache: The second use case is using the VIRTURAID controller as an in-server cache device. In this mode, hundreds of TeraBytes of NVME drives, on-board GigaBytes of RAM, and TeraBytes of in-server RAM are combined together to act as a cache device for any centralized storage connected.
The fact that large amounts of ‘hot’ data is now on the locally attached RAID controller, on the same motherboard as the CPU processing/utilizing this data, improves throughput and latencies, compared with cache in centralized storage appliances, where ‘hot’ data needs to traverse the network and appliance controller.