Drilldown on recent processor activity as measured by the Windows server.
To see the CPU drilldown
Click Monitor | CPU.
To investigate... | Look at the... | |
---|---|---|
Performance details for each CPU on the Windows server |
CPU page |
Shows the CPU utilization for all the processors on the Windows server. Each processor is represented by a single line on the chart, and by an item in the legend at the top right of the chart. To view CPU utilization analyzed according to processing mode, see the Total CPU Utilization chart on the Processing page in the CPU drilldown. Shows the number of threads (program execution units) that are waiting to run on each processor. A sustained queue length greater than four may indicate processor congestion. Shows the number of interrupts per second per CPU on the Windows server. Each CPU is represented by a single graph on the chart, and by an item in the legend at the top right of the chart. The Interrupts chart is designed to show if any CPUs on the system are bearing a disproportionate level of the I/O load. Shows the number of kilobytes per second transferred per CPU on the Windows server. Each CPU is represented by a single graph on the chart, and by an item in the legend at the top right of the chart. |
Overall performance details across all CPUs on the Windows server |
Processing page |
Shows the combined CPU utilization for all processors on the Windows server. The following are displayed:
Shows the total number of threads (program execution units) that are waiting to be run on ALL processors. A sustained processor queue length greater than ten can indicate processor congestion. To view Server Work Queues for the individual CPUs in the system, see the CPU Utilization chart on the CPU page in the CPU drilldown. Shows the total number of Windows processes and threads that exist. A process is one instance of an application program or system service that is currently executing on the server. Each process will have one or more threads which are the basic entity that can be scheduled. Sophisticated application processes such as SQL Server or Exchange can have dozens of threads running concurrently. Shows the rate at which Windows has been processing context switches. A Context Switch occurs when a processor switches from one thread to another. Context switches occur when a running thread voluntarily relinquishes the processor, is preempted by a higher priority ready thread, or switches between user-mode and privileged (kernel) mode to use an Executive or subsystem service. |
Virtualized CPU page |
VMware server host Shows usage statistics for the CPU of the virtual machine as percentages of the theoretical maximum CPU available to the virtual machine.
A low idle percentage indicates that the virtual machine may not be getting the CPU time it requires. This can be seen in the chart when the overhead and used percentages meet. Shows how much CPU the virtual machine is using, in megahertz. Hyper-V server host Show CPU usage statistics.
Show the average queue time for the virtual machine waiting for CPU to become available. The average queue time should remain under 60,000ns. If the average queue time exceeds 60,000ns then an alarm is raised. A high Hyper-V CPU wait time per dispatch alarm is raised if the average queue time exceeds 100,000ns. The current value for this metric is displayed on the Windows Server | Home Page | CPU Panel Virtualization Overhead control. Show the load on the logical processor and virtual processor.
A logical processor executes items for many virtual processors. All items to be executed are placed in a queue and the logical processor executes all that is queued. |
|
TIP: For more information on Spotlight charts and grids see Spotlight charts and Spotlight grids |
© 2022 Quest Software Inc. ALL RIGHTS RESERVED. Feedback Terms of Use Privacy