To sample the scheduler queue, read the runnable counts of every scheduler eight times, one second apart, and compare the result with a baseline. This post shows the queue on a quiet server and again while more busy sessions than schedulers run. It reads the spread, the streak and the average that tell pressure from noise.
Scheduler ID Values in dm_os_schedulers: What the Big Numbers Mean
Scheduler ID values in dm_os_schedulers run from 0 up to a few thousand, and then jump past one million. This post groups the rows into user, dedicated admin connection and hidden schedulers, and shows the filter that keeps CPU scripts honest. It also shows why filtering on the number alone is fragile.
Choosing a Processor for SQL Server: Count the Licensed Cores
Choosing a processor for SQL Server starts with the license, because SQL Server is licensed per core and the cores you buy decide the bill. This post reads what the server reports, counts the cores that matter, and explains the edition limits and the per-core speed trade. It ends with a way to size cores from measured CPU use.
Reasons for Slow Performance in SQL Server: The Top Five
The reasons for slow performance in SQL Server fall into five groups: schema design, T-SQL code, indexing, server deployment and hardware. This post tests three of them on a demo database, with logical reads as proof. It also shows a query that checks four deployment settings.
What is the Priority of Database Scoped Configurations? – Interview Question of the Week #254
Question: What is the Priority of Database Scoped Configurations over Global Configuration at Server Level?
Answer: This is a follow-up question.
Task Manager Update Speed: Why the CPU Graph Looks Stuck
Task Manager update speed controls how often the CPU graph moves. When it is set to Paused, the graph freezes and a normal server can look like it has a CPU problem. This post shows the fix and a five second test in T-SQL that tells you whether the graph is alive.
Finding Regressed Queries in Query Store After a Deployment
Find regressed queries in Query Store using weighted before-and-after duration and CPU, then compare plans and changed procedure text.