Administrator Guide

Live Volume support for Microsoft Windows/Hyper-V
70 Dell EMC SC Series: Synchronous Replication and Live Volume | CML1064
9 Live Volume support for Microsoft Windows/Hyper-V
This section details aspects of the Live Volume feature set that are specific to Microsoft environments, such
as best practices for MPIO settings and Live Volume automatic failover (LV-AFO), including ALUA support
with SCOS 7.3. It is recommended to review the prior general sections of this document before proceeding in
this section. In addition, review the VMware sections because Live Volume and MPIO behavior are very
similar in both environments.
9.1 MPIO
Microsoft Windows or Hyper-V servers running 2008/R2 and newer versions on SC Series arrays can use the
in-box Microsoft MPIO device specific module (DSM). The DSM comes with different MPIO policy options.
The following policies are supported with Live Volume:
For SCOS 7.2 and prior:
• Round Robin (no ALUA support; all paths reported as optimized)
• Failover Only
For SCOS 7.3 and newer:
• Round Robin
• Failover Only
• Round Robin with Subset, Live Volume ALUA support (see limitations described in sections 9.3 and
9.4)
Note: For more information about Windows Server and MPIO policies, settings, and best practices, review
the Dell EMC SC Series Storage and Microsoft Multipath I/O best practices guide.
9.2 Round Robin
The Round Robin policy spreads the read and write I/O evenly over all available data paths to a data volume
without any consideration for latency or bandwidth for individual data paths. All available data paths are
treated as optimal. Typically, this unintelligent yet simple approach is an ideal configuration in the following
scenarios:
• If using non-uniform server mappings (where a host is presented with multiple data paths only to the
SC Series array that is local to it)
• When the workload is local to the SC Series array that hosts the primary Live Volume(s)
• If bandwidth and latency performance are the same for all the available data paths
Round Robin also works well with uniform server mappings in the following cases:
• When both primary and secondary Live Volumes are accessible by high-bandwidth low-latency
primary and secondary data paths.
• If the slight latency penalty caused by proxied data over secondary data paths does not negatively
impact the workload.
In many cases, using non-uniform server mappings with Round Robin provides the best overall design while
minimizing complexity, and it provides a good design baseline.