Install Guide

11 Best Practices for Sharing an iSCSI SAN Infrastructure with Dell PS Series and SC Series Storage using VMware vSphere
Hosts | 2015-A-BP-INF
Note: Additional PS Series-specific information can be found in ESXi Versions 5.1, 5.5, or 6.0 Host
Configuration Guide and Dell PS Series Configuration Guide.
3.4 SC Series specific settings
A typical SC Series iSCSI implementation involves two separate, dedicated Ethernet fabrics as two fault
domains with an independent IP subnet and unique, non-default VLANs in each switch fabric. However, to
enable SC Series to coexist with PS Series and share the Ethernet SAN infrastructure using the iSCSI
storage protocol, use a single subnet for all host and storage ports.
To implement this correctly, a basic understanding of PS Series and SC Series storage is needed. This paper
provides an overview of both storage types.
Each PS Series volume is presented as a unique target with LUN 0. The PS Series volumes that are
accessible to the host are listed in the iSCSI initiator properties. When a volume is connected, the iSCSI
initiator establishes the initial iSCSI session and then the PS Series MPIO plugin determines if additional
sessions are necessary for redundancy.
Each SC Series array has both front-end and back-end ports. The front-end ports are presented with unique
target LUN IDs. Every initiator IP has a connection to each port that it can access. Redundant connections
are made by creating multiple sessions with each of the virtual iSCSI ports of the SC Series storage system.
For example, one initiator port and two target ports in each fault domain means there will be four connections
(two for each fault domain).
Note that the host port on one fault domain can access the target port on the other fault domain through the
switch interconnection. Ensure that the host ports are connected to the appropriate fault domain and target
port, physically and in iSCSI sessions. This minimizes inter-switch link (ISL) traffic and ensures that at least
some iSCSI sessions will persist in the event that a component fails in a fault domain. The following sections
discuss ways to ensure that the physical connectivity and iSCSI sessions are established correctly.
Note: Additional SC Series-specific information can be found in Dell EMC SC Series Best Practices with
VMware vSphere 5.x-6.x.
3.4.1 SC Series host physical connectivity and IP assignment
Depending on the OS-specific implementation, different methods are used to connect the arrays and assign
IP addresses. Since SC Series fault domains are connected by an ISL and are in a single IP subnet, it is
important to ensure that iSCSI sessions are properly established within their fault domains. Host ports
connected to Fault Domain 1 should connect to switch fabric and storage ports on Fault Domain 1 physically.
The same rule applies for Fault Domain 2. This step is important because, with a single subnet, it is possible
for the hosts to access SC Series storage ports on both fault domains. The correct connectivity minimizes ISL
traffic and ensures that at least some iSCSI sessions will persist in the event of a component failure.