Home > Storage > ObjectScale and ECS > Product Documentation > Dell ObjectScale: Overview and Architecture > Erasure coding
ObjectScale protects data using Dell object storage erasure coding (EC) mechanics. EC is a method of data protection in which data is broken into segments, expanded, and encoded with redundant data segments, and stored at various locations across the storage platform. ObjectScale implements a Reed Solomon error correction scheme for production environments:
12+4—Chunk is broken into 12 data segments, and four coding (parity) segments are created. A minimum of five disks are required per node for a four-node cluster with 12+4 EC scheme. This allows for one disk failure from a single node or a single node failure.
For the EC scheme, the resulting data and coding segments of each chunk are equally distributed across the nodes in the Kubernetes cluster.
Upon a Kubernetes node permanent failure, the copies of lost data segments are re-created using remaining data and coding segments. During temporary Kubernetes node failure, data services continue with data and coding segments that are being used to re-create data when needed.
ObjectScale minimum disk requirements vary based on object store EC requirements. When an object store is created, the total raw capacity and EC scheme are specified. Administrators choose the topology based on input to provide optimal protection and SS size. The number and size of SS instances in an object store represent the persistent storage capacity allocated for raw user data. SS instances attach to Kubernetes persistent volumes (PVs) on disks using Kubernetes persistent volume claims (PVCs). ObjectScale writes data for best protection considering number of volumes on disk, disks per SS, and SS instances across the cluster.
During the object store creation process, the number of Kubernetes nodes, either physical servers or worker nodes, in the cluster determine the available EC schemes that are presented within the New Object Store wizard. ObjectScale uses the Kubernetes anti-affinity rules to ensure that the SS replicas are properly placed across the nodes in the cluster. The New Object Store wizard ensures that the number of SS replicas for the new object store is not below the minimum for the selected EC scheme.
The EC scheme of an object store cannot be changed after deployment. Capacity may be added to an object store, but the EC scheme does not change, only the amount of available storage changes.
The following table lists the available EC data protection schemes in ObjectScale:
EC scheme (data+coding segments) | Minimum nodes required | Data availability during component failures |
12+4 | 4-5 | 1 node failure or |
12+4 | 6-9 | 1 node failure or |
12+4 | >=10 | 2 node failures or Disk failures from up to two different nodes and up to a maximum of four disk failures in total
|
For more details about EC in Dell storage, see ECS: Overview and Architecture.
The number of replicas and volumes per SS may be increased in an object store using horizontal or vertical expansion:
For more details about EC in Dell storage, see ECS: Overview and Architecture.