In this one, we will compare the two portals in terms of application-level dashboards.
If you’re not familiar with Azure’s Application Insights service, or would like to learn more about Azure Monitor in general, I’d encourage you to check out the Azure Monitor Learning Path series that we recently concluded.
A couple of pre-requisites for this task:
- On-board your application into Azure Monitor with Application Insights. More info on that here.
- Enable any solutions that are useful to measure the performance of your application. More on that here.
Right, onto the task at hand. Let’s make that dashboard!
Task: Create a dashboard to provide all-round visibility of my applications hosted in Azure AppInsights.
Let’s start by entering into the Azure portal and browsing to our application in Azure Monitor. We’ll work with the “Parts Unlimited” sample application.
First off, I see there is a standard dashboard already built for me by Application Insights. Let’s take a look.
This is what it looks like:
Hmm, looks good. I’ve got some key metrics to start off. Like any standard dashboard in Azure, you can add more/remove tiles according to your requirements. To learn more about adding visualizations to your dashboard in Azure, see the “Visualizing Log Analytics Data in Azure” part of the AMLP series. I’m aware it says “Log Analytics data” in the title, but it is largely the same process for AppInsights.
On top of this, I can quickly pin the cost dashboard scoped to this resource from the Cost Management service, which looks like this:
Right, so this is good and I can add more KPI’s using KQL queries, but still there is some data I’d like to see along with it, like some external data I’ve integrated my application with and so on.
- A fairly useful standard dashboard to start with
- Limited visualizations
- Customization could’ve been better
- Some key aspects of application missing
- Many visualizations are simply “shortcuts” into the resource
Let’s jump into the SquaredUp equivalent and see how that fares.
In the SquaredUp portal, hit the + sign to make a new dashboard. As always, I’m presented with different tile options.
I won’t talk about the detailed process of adding tiles to the dashboards here, as I’ve already done it in Part 1 of this series. I’d encourage you to take a look at it to learn more about tiles, scoping and customization.
Right, so after adding a few of the tiles that interest me, I’ve made a dashboard like this:
Note the areas highlighted in red.
These are some of the tiles I’ve added very easily in SquaredUp against my application. Along with the common performance data, these show me how much the application is costing (Cost tile), its performance in terms of connectivity to my remote users (Web API tile), what the actual components that make up this application are (Resources tile), as well as if there are any alerts from this specific application (Alert tile).
A quick look at the Azure native dashboard shows that it is missing this visibility.
Another interesting set of data is visible if I go one step up from here – an overview of all my applications.
As you can see, this is a high-level overview of some of my application. This includes a LIVE preview of my applications, the cost of the applications and some of the external data related to individual applications, like the Service Now incidents and response times using Pingdom. There’s a good chance you are monitoring the availability and/or performance of your applications using tools outside of Azure. You can easily pull that data into SquaredUp using their APIs, which means the list of external tools you can use is practically endless. Anything with a JSON API can be brought into SquaredUp. This capability is simply missing in the Azure portal.
- Easy to customize and configure
- More visualizations available
- Easy visibility from external data sources
- Drill down capability
- True single pane of glass for complete visibility into my applications!
Alright, that’s it on this one. Let’s meet again in the next (and final) part to see how the two dashboards stack up for the most requested data visualization – cost analysis!