Modern web applications operate on a scale that makes manual observations ineffective in their maintenance. Beyond that users have an expectation of continuous access and functionality. As a system administrator you will use tools like Azure Application Insights to produce metrics that will make it possible to aggregate real time health and performance data of the web application. From these metrics charts can be created that will allow analysts to keep a wider view of various key metrics for even the largest application.
Configure Application Insights:
For every application created in Azure it is possible to enable a metrics collection called Application Insights. Its main job is to collect information that is valuable to evaluate whether the application is working properly and what its usage stats. Azure has made it simple to create a configuration for the metrics collection from within the portal.
Analyze data by Using Application Insights:
The data collected by Application Insights would be useless if the only readout were tables or logs. Analysts must be able to quickly follow key stats that give an overall view of how the application is running. Any good analytics platform will provide visualizations in the form of various graphs and charts that provide real time aggregations. Dashboards can then be built that allow analysts to “drill down” into the data when they find abnormalities or other points of interest. Notice how easy it is to pivot from one set up metrics to another and change the time period being viewed. This is how system administrators make up for the massive scale of most web applications and still maintain satisfactory availability.
Test Application Availability:
The Azure Application Insights service provides more than just passive metrics. It also has the ability to make various tests that will actively trigger behavior from the application. Depending on the type of information needed the tests can take many different forms. The test that this particular section requires is a simple availability test. However do not underestimate its importance. Imagine if admins had to check pages of logs each time they wanted to know if an application was running properly. This would not work at all and would eventually bring the service to failure. Insight’s ability to perform regular availability tests literally makes an admin’s job even possible.
Lab Summary Conclusion:
Users expect applications to be available 99% of the time, and so do the companies that pay for their operation. System administrators face a daunting challenge with the scale available to developers with cloud computing. To assist admins Azure offers the Application Insights service out of the box. It delivers priceless information about the functionality of the application, and can provide active testing for the most important datasets. However all this data would be useless without the ability to visualize with graphs and charts. Metrics is a growing field important for any system administrator to be able to deploy for any application they oversee.