Specialized Performance Troubleshooting (Part 3: How to identify storage issues at a SQL Server box)

This week I will explain how to detect and analyze storage performance issues. The first step is to collect a Perfmon trace and check these two counters for each disk: Physical Disk: Avg. Disk Sec/Read Physical Disk: Avg. Disk Sec/Read The recorded values should be less than 20 msec for attaining good performance. They will have to remain below 40 msec at all costs, otherwise performance can get impacted. The next step is to find out which process causes this IO. So we will record values...(read more)