Administrator Guide

Live Volume overview
41 Dell EMC SC Series: Synchronous Replication and Live Volume | CML1064
It is recommended to use dedicated VLANs or fabrics to isolate IP-based storage traffic from other types of
general-purpose LAN traffic, especially when spanning data centers. While this is not a requirement for Live
Volume, it is a general best practice for IP-based storage.
For hypervisor virtualization products such as VMware vSphere, Microsoft Hyper-V, and Citrix
®
XenServer
®
, a
site must have at least a 1 GB connection with 10 ms or less latency between servers to support VMware
vMotion
®
Metro or live migration activities. Standard vMotion requires 5 ms or less latency between the
source and destination host.
5.4.1 High bandwidth, low latency
For an inter-data-center, campus environment (or within a 60-mile radius as an example), high-speed fiber
connectivity is possible. While inter-data-center and campus environments may be able to run fiber speeds of
up to 16 Gb using Multi-mode fiber connectivity, single-mode fiber connectivity of up to 1 Gb using dark fibre
can assist with connecting data centers together that may be up to 60 miles apart as an example. Minimal
latency is especially key when implementing Live Volume on top of synchronous replication (in either mode).
This type of connectivity is highly recommended for live migrating virtual machine workloads between arrays
and required for synchronous Live Volume with automatic failover used in vSphere Metro Storage Cluster
configurations or Microsoft Windows Server/Hyper-V clusters.
5.4.2 Low bandwidth, high latency
If a site is planning on using Live Volume on a low-bandwidth, high-latency replication link, it is recommended
to control swap role activities manually by shutting down the application running at site A, perform a Live
Volume swap role, and then bring the application up at the remote site. This scenario prevents any storage
proxy traffic from going across the link, as well as providing a pause in replication I/O for the link allowing the
replication to catch up so that a Live Volume swap role can occur as quickly as possible. Manual swap role
activities can be controlled by deselecting the Automatically Swap Roles option on the Live Volume
configuration. If the replication connection has characteristics of high latency, asynchronous replication is
recommended for Live Volume so that applications are not adversely impacted by replication latency. Note
that Live Volume automatic failover supports synchronous replication in high availability mode only.