The difference between application monitoring and infrastructure monitoring and why both are necessary to avoid downtime of your ERP. Christer Wecke breaks down the two concepts in this blog post and has written an in-depth guide that you can download here.
Different types of monitoring
In our industry, we often talk about two types of monitoring; Infrastructure monitoring and Application monitoring. Unfortunately, these two types are frequently mixed up. As many hosting providers offer monitoring as part of their standard services it is often believed that business systems are monitored (with application monitoring) when, in fact, there are only certain system services in the business system that are being monitored.
The day one becomes aware of application monitoring is often the day after experiencing a situation. A typical situation can be that your colleagues simply cannot work whilst the IT department frantically is searching for the cause of the production shutdown.
Today's applications often consist of many different parts spread across different servers, linked by a network of multiple switches and firewalls. The most common way to keep track of all of these parts is to use a monitoring solution that keeps track of an IP address always responding or a specific service is always available on the network, so-called Infrastructure Monitoring.
This is fine; however, this kind of monitoring does not answer the question if the application and the necessary processes is actually working to 100%.
Today, with more and more EDI connections to the ERP, it becomes even more important to track all processes and quickly identify the processes that are not working in a satisfactory way.
Once you've identified where it has gone wrong, you will need to find someone who can fix the error. A common scenario is when a number series is about to end - who keeps track of this and fixes it before billing? Or having a table containing errors but still being replicated to your data warehouse, causing the numbers in the Business Intelligence system to be incorrect?
The above are just a few examples of when traditional monitoring or infrastructure monitoring is insufficient.
For successful monitoring: understand your application AND your processes
To get started with your application monitoring, you need to understand how your application is set up so that you fully understand the functionality, database structure and what kind of information is available in the different tables. Not forgetting, most importantly - you must understand how the application supports your business processes.
Once you check these sections, you can start designing your application monitoring and control points that you need to make sure the application is up and accessible to all users without any interference.
To sum up, monitoring mainly deals with three things:
To ensure that business-critical systems work as they should.
When possible; to predict the problem before it happens.
To find out the cause of a malfunction directly or as quick as possible
Want to know more? I have written a guide where I develop the benefits of application monitoring. Download it below.