White Papers

Administration best practices
22 Dell EMC SC Series: Microsoft Windows Server Best Practices | 680-042-007
SC Series volume snapshots
For guidance on how to configure Windows Server and Hyper-V with Live Volume, see the document Dell
EMC SC Series Storage: Synchronous Replication and Live Volume.
3.10.1 Crash-consistent and application-consistent snapshots
When a snapshot is created of a volume on an SC Series array according to a schedule (or on demand)
without regard to the state of the server, its applications, or data, it is assumed to be crash-consistent.
When recovering a server (or data) using a crash-consistent snapshot, it is like recovering from a power
outage at that point in time. In most cases, servers and applications are resilient enough to recover to a
crash-consistent state without any complications, whether the cause is an unexpected power outage, or the
server is being recovered to a previous point in time due to an event such as a malware infection.
An exception to this is when the environment hosts a transactional workload such as Microsoft Exchange or
SQL Server
®
. With transactional workloads, the risk of data loss or corruption is higher when attempting to
recover to a crash-consistent state, especially when databases span multiple volumes.
Crash-consistent snapshots should not be relied upon for recovery when application consistency is critical for
recovery of the workload.
There are several ways to ensure application-consistency. For example, an administrator can power off a
server and its workload, or pause an application temporarily, before taking snapshots. However, manual
processes may not be very practical and risk introducing user error. This can be mitigated in cases where
automation can be used to script processes, so they can, for example, run unattended after business hours.
See section 4.9 for more information on leveraging the PowerShell SDK toolkit with SC Series storage.