Dell PowerFlex Appliance and PowerFlex Rack with PowerFlex 3.x VMware NSX Ready Architecture Overview

PDF

System overview

This section describes the PowerFlex rack and PowerFlex appliance overview for VMware NSX ready.

The following table shows the hardware aspects that can be customized in a base configuration for PowerFlex rack and PowerFlex appliance:


Hardware	PowerFlex rack	PowerFlex appliance
Management	Dell requires three PowerFlex management controller nodes with high availability (HA) support and vSAN for software scale-out. To support VMware NSX Edge nodes, a fourth node is required to support the required system resources to support the NSX Manager VMs.	PowerFlex management controller nodes are either part of the appliance or provided by the customer. Two options are available: A single node or a three-node PowerFlex management controller. Neither option meets the VMware NSX requirements. To support NSX requirements, four PowerFlex management controller nodes are recommended to support the required system resources for three large-size NSX Manager VMs. NOTE: If the PowerFlex management controllers are provided by the customer, the customer must review the NSX documentation provided by VMware, and provide sufficient infrastructure to support the required CPU and memory for the NSX Manager VMs.
Compute	Dell Technologies recommends at least four PowerFlex hyperconverged nodes or three PowerFlex compute-only nodes for the performance and resiliency features of storage data server (SDS).	Dell Technologies recommends at least four PowerFlex hyperconverged nodes or three PowerFlex compute-only nodes for the performance and resiliency features of storage data server (SDS).
Network	Supports either a traditional Ethernet architecture (Cisco or Dell Technologies) or leaf-spine topology (Cisco). By default, the NSX Edge physical nodes connect directly to either the aggregation or border leaf switches, depending on the network topology. If there is a limitation because of port capacity or cable distance, the management and transport connections are relocated from the aggregation or border leaf switches to the access or leaf switches.	The access switches (Cisco or Dell Technologies) can be managed by PowerFlex Manager or managed by the customer. The leaf (Cisco), spine (Cisco) or aggregation (Cisco or Dell Technologies) switches are only managed by the customer. Dell Technologies recommends connecting the customer provided NSX Edge nodes directly to the border leaf or aggregation switches. If there is a limitation because of port capacity or cable distance, the management and transport connections are relocated from the aggregation or border leaf switches to the access or leaf switches.
Storage	PowerFlex storage-only nodes are not supported as an NSX transport node. Storage is software-defined SAN storage that uses local disks to build a virtual SAN storage pool. PowerFlex management controller nodes run on VMware vSAN as the storage solution. PowerFlex compute-only nodes run on PowerFlex as the storage solution. NSX Edge nodes can run in either a local RAID storage (recommended) or a VMware vSAN storage solution.	PowerFlex storage-only nodes are not supported as an NSX transport node. Storage is software-defined SAN storage that uses local disks to build a virtual SAN storage pool. The single PowerFlex management controller node runs on local RAID storage as the storage solution. A single PowerFlex management controller node cannot be used for NSX. The three PowerFlex management controller node option runs on vSAN storage solution. NSX requires a fourth PowerFlex management controller node to meet system requirements. PowerFlex compute-only nodes run on PowerFlex as the storage solution. The NSX Edge nodes can run in either a local RAID storage (recommended) or a VMware vSAN storage solution.

These resources can be scaled up to meet application workload requirements by adding PowerFlex nodes to the environment.