PowerFlex is designed with a scale-out architecture, allowing for seamless expansion of resources. The primary method for increasing the total storage capacity of a PowerFlex cluster is through Dynamic Node Addition. This involves adding new Storage Data Server (SDS) nodes to the existing cluster. Once a new SDS is added, its storage devices are automatically discovered and integrated into the storage pool. PowerFlex then initiates a rebalancing process to distribute data across all nodes, including the new one, ensuring balanced capacity utilization and performance without requiring any downtime. This elastic scalability is a core tenet of the PowerFlex platform.

A. Fault Sets: This feature is for high availability. It groups nodes into failure domains to ensure data replicas are not placed on servers that might fail concurrently (e.g., in the same rack).

C. Snapshots: This is a data protection feature that creates point-in-time, space-efficient copies of volumes for backup and recovery purposes. Snapshots consume capacity, they do not add it.

D. Shared File Systems: This refers to the capability of PowerFlex to provide file services (NFS/SMB) on top of its block storage foundation. It is a use case, not a mechanism for capacity expansion.

1. Dell PowerFlex: Introduction and Concepts Guide (Version 4.5)

Page 11

Section: Elasticity and scalability: "PowerFlex is a scale-out system that can grow from a few nodes to thousands of nodes. You can add or remove nodes on the fly

without any downtime. When you add nodes

the system automatically rebalances the data to take advantage of the new resources." This directly supports Dynamic Node Addition for capacity expansion.

Page 20

Section: Fault sets: "A fault set is a group of SDSs that are configured so that they are not likely to fail at the same time... Fault sets are used to increase the availability of the system." This confirms its purpose is availability

not expansion.

2. Dell PowerFlex: Architecture Guide (Version 4.x)

Section 2.2

Scalability: "PowerFlex can scale out by adding more servers to the cluster. When a new server is added

its resources are automatically incorporated into the system

and the workload is rebalanced across all servers." This reinforces the concept of adding nodes to scale.

3. Dell PowerFlex: Best Practices Guide

Section: Scaling the PowerFlex cluster: "To scale storage capacity and performance

add SDS nodes to the protection domain. PowerFlex automatically detects the new resources and rebalances the existing data to make use of the new capacity and performance." This explicitly links adding SDS nodes to scaling capacity.