Embedded Files
Performance Monitoring
Performance Monitoring
Keep in mind, just because you can do something does not mean you should always do it. This holds true with production system. You should not make causal changes to a production system!
vCenter Server provides amazing performance charts that we can use to help understand how our systems are performing. You can view data on: clusters, hosts, datastores, networks and Virtual Machines. The overview performance chart displays the most common metrics and is a great way to gain a nice overview of what is going on with your system. See image below for an example of the Overview Performance Chart. You should be using these performance charts to view and improve performance of your VMs.
Chart Options
Chart Options
You might notice that vCenter Server stores data/statistics at different time intervals. Please reference the table below to understand what each of these time intervals means and how often the data is collected. Depending on what you are data you are monitoring you might be presented with either a bar, pie, line or stacked chart. You can find a sample of each chart type below.
Line Chart
Line Chart
Stacked Chart
Stacked Chart
To create charts based on custom data is possible under the Performance --> Advanced section under the Monitor tab. You must select the type of data you wish to view and the chart type.
You can also view information about the host under the Utilization section. This shows you data about the Host CPU, Host Memory and Guest Memory (your VMs). This can help you gain an understanding of the performance of your overall server and with each VM. As you can see my server has a lot of available resources.
Saving Charts
Saving Charts
It is possible to save a chart. You just need to click on the Save Chart icon (top right of chart) to save it. You can then save the data in various formates to meet your needs.
Hardware Health
Hardware Health
Not only do we have to worry about how our VMs and the host is performing, but we should also be monitoring our hardware health. This can help determine if we might have a hardware failure in the near future and/or steps we should be taking if that is the case. Using the vSphere Client you can monitor a variety of hardware such as:
CPU processors
Memory
Fans
Temperature
Network
Battery
Storage
PCI Device
And more!
To monitor the hardware health make sure you have a host selected in the vSphere Client, click the monitor tab and then click Hardware Health. You can then select what you wish to view.
Monitoring Events and Alarms
Monitoring Events and Alarms
What are events? Events are actions that a user or the system performed on a host. Examples of events are that a license key has expired or that a VM was powered on. These are important as they provide an insight into what has been taking place on your servers. There are also different types of events.
Error: A fatal problem has happened and the process or operation was stopped.
Warning: A potential risk exists and it should be fixed ASAP.
Information: Lets you know that an operation was done successfully.
Audit: Provides details about who, when and what was done.
What are alarms? An event could trigger an alarm or a predefined condition could also trigger an alarm. Alarms contain the following information:
Name and description of the alarm.
What is being monitored.
Alarm Rules - defines what the conditions or event that would trigger an alarm is and it also states what should be done in response to the alarm being triggered.
Timestamp of when the alarm was last modified.
An example of an alarm that can be configured is the host CPU usage. You might want to enable this alarm for when a host reaches a predetermined % of CPU usage. Another example of an alarm that you might want to use is monitor your datastore disk usage. You might want to trigger an alarm when the datastore reaches 70% allocated.
Page updated
Google Sites
Report abuse