Introduced in SQL Server 2012, Always On Availability Groups maximizes the readiness of a set of user databases as an availability group that can fail over as a unit. In this session we will discuss the improvements available in SQL Server 2017 and how these improvements increase the feasibility to leverage AlwaysOn Availability groups as a High Availability / Disaster Recovery solution for reporting and OLTP workloads across Windows and Linux environments. We will demo SQL Server 2016 use cases of ‘increased number of secondaries’ and other key capabilities. We will cover SQL Server 2016 specific capabilities such as improved log performance, multiple replica automatic-failover, failover based on database failure, readable secondaries for ‘round robin’ load-balancing, and basic AlwaysOn ‘Availability Groups’ for Standard edition. Finally we will discuss SQL Server 2017 on Linux using Pacemaker clustering and read-scale Availability Group deployments. In SQL Server 2017 there are now three different architectures for availability groups. Always On availability groups can now provide high availability, disaster recovery, and read-scale balancing. In Windows, failover clustering provides the cluster manager while in Linux, you can use Pacemaker. The other new architecture is a cluster-less, read-scale availability group. We will discuss how a read scale availability group provides support for non-HADR, read-only workloads.
For the past nine years, I have been a Technical Solution Professional with Microsoft. As a TSP I conduct full training workshops, conduct demo sessions, and perform architectural reviews on all supported builds of SQL Server. I mainly support the Southeast focusing on Florida, but I can be anywhere in the world. My core areas are Performance Tuning and Optimization, Azure, AlwaysOn, Reporting Services (SSRS), custom monitoring, building custom dashboards, and more.