Monitoring systems consist of interconnected components across the ecosystem. They can be broadly categorized into three key parts. Now, let’s delve into each of these primary “3 layers” and explore their significance.

The Foundation Layer
This layer, which forms the basis for advanced monitoring capabilities, involves monitoring physical or virtual devices known as ‘hosts.’ These hosts encompass a wide range, including Windows and Linux servers, Cisco routers, Nokia firewalls, and VMware virtual machines.

The foundational layer focuses on ensuring these hosts are operational by sending ping requests. Once configured, this layer provides a view of the added hosts, indicating which ones are up or down. This basic information serves as the foundation upon which advanced monitoring is built.

The IT Monitoring Layer
Beyond the foundation layer, this layer delves into the monitoring of specific items running on these hosts. For instance:

• On Linux servers: Swap space, running services, CPU usage, etc.
• On Windows servers: Pagefile size, CPU utilization, memory usage, available storage on C:/, running processes, etc.
• For virtualization (e.g., VMware): Datastore availability, temperature, the number of virtual machines, CPU utilization, etc.

These monitored items are referred to as “service checks,” and they are executed on the hosts specified in the foundation layer. The process essentially involves examining the performance metrics of these items. Innovations in monitoring have led to the development of “Autodiscovery.” This feature enables monitoring systems to scan and discover devices within predefined subnets or networks.

For instance, in the case of Windows servers, scanning a subnet allows the system to discover and import all hosts on that network. The monitoring system can also determine the operating system of these hosts and automatically apply templates based on the results, ensuring a swift time-to-value.

The Interpretation Layer

Now that the monitoring system is tracking the health and performance of hosts and the services they run, it’s time to interpret the data intelligently. This involves answering questions like, “How can we present this data in a way that highlights issues clearly?”.

In IT, servers and network devices come together to form larger objects such as applications, websites, or web services. The primary focus should be on monitoring these larger entities rather than their components. After all, the ultimate concern is the impact of IT issues on the business and its customers. To address this, monitoring software vendors have introduced “business service monitoring.”

Business service monitoring allows users to gain insights into the performance of applications, stacks, websites, and other complex entities. It focuses on their health as a whole rather than the status of their components. It provides a “top-down view” of services, prioritizing the impact on business services rather than examining the underlying components.