Monitoring SQL Servers in the enterprise can be complex and it is difficult to get a real assessment of a single SQL Server instance and much more difficult to get the right performance health assessment from multiple servers in your enterprise.

Additionally, performance monitor counters aren't clear which ones to use and how to match these performance counters to DMV and wait statistics outputs.

In this session, we step way outside the box. We'll start with how Microsoft itself collects diagnostic data using SQL Log Scout, SQLdiag/PSSDIAG, and the PAL tool for threshold analysis these are the same tools CSS support engineers use when your escalation lands on their desk.

From there, we dive deep into PerfMon composite templates as deployable code, building your own custom performance counters using the SQL Server: User Settable Object counters, leveraging Resource Governor as a pure monitoring engine, automating alerting pipelines with PowerShell, and mining SQL Server's built-in black box - Ring Buffers and the always-on system_health session.

Every topic includes real demos and scripts you can take home and use Monday morning.

This is one of the sessions I can guarantee you will learn something new and valuable as we will be touching on some unique approaches and techniques for SQL Server performance monitoring.

You will learn:

• How Microsoft collects and analyzes diagnostic data using SQL Log Scout, SQLdiag/PSSDIAG, and the Performance Analysis of Logs (PAL) tool — and how you can use these same tools proactively before issues escalate to support.
• How to build and deploy PerfMon monitoring templates and learn how to expand the approach as code using logman, enabling consistent, version-controlled, fleet-wide performance baselines deployable to any number of SQL Servers.
• How to use SQL Server's User Settable Object to turn any query result, business metric, or SQL Server condition into a native PerfMon counter along with Resource Governor classifications for a custom monitoring solution.