Self-healing storage cluster

All data is stored in multiple copies and possibly geographically separated locations within the cluster. When a disk or a storage node breaks down, the cluster self-heals by recreating the missing data from the damaged node in the free space in the cluster without almost none capacity decrease.

Compuverde supports copies of data as well as erasure coding. Erasure coding provide you with a much better data footprint and redundacy. And the more nodes that is added to the cluster more erasure coding options gets available.

When it comes to the hardware perspective there is certain levels of failures that we can tolerate within a cluster.

Tolerated hardware failure:

  • Server
  • Magnetic disk
  • Flash disk
  • Controller card
  • Network interface
  • Network access

Failure of any hardware listed above doesnt affect the access of the data stored in the cluster. The data is distributed over all nodes and each node is a gateway to the cluster and shows the same view.


In short, our software solution allows for:

  • Telecom graded access and availability
  • Always on - possible to both read and write whenever
  • Extreme redundancy and recovery process
  • Independent of traffic volumes or file sizes
  • Constant backup and recovery - regardless of write failures

Read more about our scale-out storage solutions?