Azure API Management

Inside scoop from the API Management team

Power BI solution template

Today we are announcing the availability of the Azure API Management Analytics Power BI Solution Template. This template allows you to create eye catching, customizable business dashboards, illustrating traffic flowing through your API Management instance. Share those reports with business stakeholders, product managers and senior leadership. Create your own store of API data that you can analyze, warehouse or merge with other sources.

The solution template can be accessed at http://aka.ms/apimpbi. You will need to log in to the website with credentials associated with your Azure subscription. The site will deploy the template for you; there is no cost – you simply pay the Azure consumption cost (we estimate this to be under $10 a day).

So what do you actually get?

Read more about the Azure services we deploy and how they fit together in our separate deep dive post. We’ll focus here on the Power BI report we produce.

When the template has completed installing, you will be given a link to a Power BI report (you will need Power BI Desktop installed to open it). Save and Open the report – you will need to connect it to you the Azure SQL instance you created or selected during set up.

Important!You will not immediately see data in the report. Currently, we do not support importing a historical export of your data – the report will only show data from the time it is created.

Report pages

The At a glance page gives a summary of traffic today and over the last week. KPIs give you a quick view of how you are tracking.

The API Calls page gives you detailed insight into your traffic. You can slice in multiple ways to delve down into specific APIs, Subscriptions or time periods.

The Errors page gives you insight into errors from your backend services, and errors occurring in your API Management gateway.

The Call Frequency page helps you to find unseen patterns in your traffic. The graphs on the page illustrate call patterns from specific IP addresses. Look for unusual patterns in (darker areas signify higher traffic), and look for rhythmic patterns hat may indicate automated processes.

Finally, the Relationships page shows you calls that are likely to come sequentially (within a 1 second window), either overall or from a specific IP address. Use this report to understand how app developers are using your APIs, and look for optimizations (for example merging 2 highly correlated API calls into one).

We’d love to know what you think of this template – give us your feedback below, or through the normal channels.