Best practices for hardware performance monitoring
Servers often experience performance issues without warning—often resulting in costly downtime. Because your servers are critical to your IT infrastructure and overall business services, you need to continually monitor your server hardware health and performance. Server & Application Monitor gives you instant visibility to hardware health, operating system and virtual infrastructure performance across multiple vendors to include IBM® System X®, HP®, and Dell®.
Best practice 1: Monitor the CPU fan
If the CPU fan stops, the server will eventually overheat and either fail or shutdown to prevent serious damage to its components. You should monitor the CPU fan’s speed to make sure it doesn’t exceed the safe operating RPMs. Also, watch the fan’s historic RPM speed. A spike that lasts several minutes or hours indicates a serious issue (faulty AC, blocked fan vents, etc.).
Best practice 2: Monitor the power supply
To prevent a server outage or severe component damage, monitor the amperage, voltage, and wattage of the power supply according to the manufacturer’s recommended operating specifications.
Best practice 3: Monitor server temperature
Unusually high temperatures can adversely affect server performance and even cause permanent damage. Adding virtualization to your IT environment increases server density. It’s important to make sure the temperatures of both the server and virtual environment do not exceed the safe operating limit. Check the manufacturer’s specifications for the safe operating temperature.
You can use temperature measurements to plan and deploy virtual servers where appropriate. For example, you should not deploy a high-load Web or database server on a physical server that is approaching its temperature limit.
In addition to the server temperature, monitor the temperature, air flow, and humidity of the physical space (datacenter, closet, etc.) where your server is housed. Faulty AC, improper air flow, and dangerous humidity levels can trigger server performance issues and operational failures.
Best practice 4: Monitor server resource capacity
In addition to monitoring server hardware health, watch your CPU, memory, and disk capacity. With a virtualized server, you need to understand the CPU, memory usage, and the load for the physical host of each virtual machine.
Use SolarWinds® Server & Application Monitor to view all of your server performance metrics. This way you can proactively monitor your environment and catch issues before they become major problems.