Dell PowerFlex 4.6.x Technical Overview

Fault sets

A fault set is a logical entity that contains a group of SDSs within a protection domain.

In a large cluster that spans across several racks, a fault set can be used as a means to increase the overall cluster resiliency. If one of the racks incurs power loss the use of a fault set prevents a denial of availability. By grouping SDSs in different fault sets, a verification process occurs so that two copies of data do not reside in the same fault set which increases the cluster resiliency. When the data is written to the SDS device, two copies on two SDS in different fault sets are verified.

When defining fault sets, the term fault units refers to either a fault set, or an SDS not associated with a fault set.

There must be enough capacity within at least three fault units to enable mirroring.

If fault sets are defined, you can use any combination of fault units, for example:

• SDS1, SDS2, SDS3
• FS1, SDS1, SDS2
• FS1, FS2, SDS1
• FS1, FS2, FS3

In the following figure, there are three protection domains. The middle protection domain (fully depicted) consists of seven SDSs, each with two storage devices.

Complete the following steps to use fault sets:

1. Ensure that a protection domain exists, or add a new one.
2. Ensure that a storage pool and fault sets (a minimum of three fault units) exist, or add new ones.
3. Add the SDS, designating the protection domain and fault set, and simultaneously, adding the SDS devices into a storage pool.

   The automated deployment and installation tools follow this order automatically.

   You can only create and configure fault sets before adding SDSs to the system, and configuring them incorrectly may prevent the creation of volumes. An SDS can be added to a fault set only during the creation of the SDS.

You can also add fault sets when adding SDS nodes after initial installation.