Saves capacity by using parity in place of additional data copies; if a disk fails, the parity is used to rebuild the data. RF2 requires a minimum of four nodes, blocks or racks, and RF3 requires six.
Also called EC.
Read more: Nutanix docs
In the Ultra Transcenders books
Each book explains Erasure coding in context, with comparison tables and the common traps.
Terms in this definition
- Capacity
A reserved block of compute for a tenant whose size is fixed by its SKU; Fabric F SKUs express it in capacity units (CUs). Workspaces assigned to it unlock features like Copilot, and from F64 upwards people with free licences can view Power BI content.
- REBUILD
Drops and recreates an index to clear fragmentation. Columnstore indexes also have every delta row pushed into compressed rowgroups during this operation.
Related terms
- Data Reduction Ratio
Shows in Prism how much space erasure coding, compression and deduplication save between them. Thin provisioning and cloning are added only in the Overall Efficiency number.
- FTT
Failures to tolerate: how many disks or hosts can fail without losing a VM object, met by mirroring (RAID-1) or erasure coding (RAID-5/6). vSAN's default policy uses one.
- Inline erasure coding
Creates erasure coding strips straight away instead of waiting until data turns write cold, within one vDisk or across several. Turned on with nCLI, it is advised by Nutanix only for Object storage containers.
- Nutanix storage container
Carved logically from a storage pool, it contains vDisks and controls replication factor, erasure coding, deduplication and compression. Hyper-V mounts it as an SMB share; ESXi mounts it as an NFS datastore.
- RAID-1
Mirroring in vSAN: whole duplicate copies of an object sit on separate hosts. It handles between one and three failures, at a higher capacity cost than erasure coding.
- RAID-5
Single-parity erasure coding in vSAN, surviving one failure more economically than mirroring. ESA switches adaptively between 2+1 and 4+1, whereas OSA uses 3+1 and so needs four fault domains.
- RAID-6
Double-parity erasure coding in vSAN: data is spread 4+2 across at least six fault domains, so two simultaneous failures can be survived.
- Strip size
How much data and parity make up one erasure coding strip, written like (4,1) or (4,2). The figure comes from AOS based on your cluster; wider strips bring greater savings at a higher rebuild cost.