In this series of blog posts, you will learn in detail what goes in to SQL Server migration process. You can access the first part of "Detailed specification of the SQL Server migration planning" here.
In my previous blogpost I wrote about the Compatibility Check of the system to be migrated. In this post the topic is High Availability planning.
The major key points that affect the planning of High Availability are:
Based on the answers above, different solutions and combinations of the SQL Server HA techniques can be implemented.
In this blog post, Replication, Log Shipping and Mirroring will be skipped, because they are either not real HA mechanisms, or are deprecated features (mirroring).
The traditional HA solution has, among others, the following features and requirements:
The newer HA solution for SQL Server has at least the following features and requirements:
Virtualization is beyond the scope of this blog post. But of course, either VMware or Hyper-V Virtualization can be used in different ways with SQL Server HA Solutions.
The selected HA solution depends on many factors, but I would say that good old Failover Cluster is still the first choice for mission critical systems – especially if the hardware, including a fault tolerant Storage system, is already present. It protects the whole instance, and a lot less manual administration is needed compared to HA Groups.
HA Groups are a good choice for less expensive hardware and systems that can utilize load balancing. But this comes with the cost of more manual administration.
The next part is about phasing the SQL Server migration project.
Pekka Korhonen
Lead Architect, MCM
DB Pro Oy