CPU Co-Stop

David Klee discusses an important hypervisor-level metric:

VMware’s CPU Co-Stop metric shows you the amount of time that a parallelized request spends trying to line up the vCPU schedulers for the simultaneous execution of a task on multiple vCPUs. It’s measured in milliseconds spent in the queue per vCPU per polling interval. Higher is bad. Very bad. The operating system is constantly reviewing the running processes, and checking their runtime states. It can detect that a CPU isn’t keeping up with the others, and might actually flag a CPU is actually BAD if it can’t keep up and the difference is too great.

This is extremely useful information for DBAs in virtualized environments.  My crude and overly simplistic answer is, don’t over-book vCPUs on hosts running important VMs like your SQL Server instances.

Related Posts

Migrating Lots Of Databases To SQL Server 2016

Andy Levy has a problem. Well, about 8000 of them. In part 1, he describes the plan: How do you move eight thousand databases in a reasonable amount of time? I spent about an hour and a half one morning hashing ideas out w/ folks in the dbatools Slack channel, plus several conversations in the office and with […]

Read More

Automation With Powershell Desired State Configuration

Jess Pomfret takes us on a journey to a desired state: PowerShell DSC is a platform to support the concept of Infrastructure as Code (IaC).  It uses declarative syntax instead of the usual imperative syntax of PowerShell.  This means that you describe your desired state rather than the specific steps needed to get there.  There […]

Read More

Categories

March 2016
MTWTFSS
« Feb Apr »
 123456
78910111213
14151617181920
21222324252627
28293031