In this post I want to share the results of my initial testing of the Microsoft Azure emissions insights capability within Microsoft Fabric. Which is another capability that comes with the Sustainability data solutions in Fabric.
It allows you to store and analyze your Azure emissions data. After testing the new ESG data estate capability in my previous post I thought it was appropriate to cover this one as well.
I am glad that I get to publish this post during the same week as Earth Day. Especially when so many people think everyday should be an Earth Day.
By the end of this post, you will know more about the Microsoft Azure emissions insights capability. Plus, ways that you can view your own metrics. Which I also manage expectations about as well.
Along the way I share plenty of links.
One key point I want to highlight is that the Sustainability data solutions in Microsoft Fabric are currently in preview and are subject to change.
Setting up Microsoft Azure emissions insights
As shown in my previous post, I had already deployed the Sustainability data solution to Microsoft Fabric through the Microsoft Cloud Solution Centre.
So for this capability I started by creating a new workspace in Microsoft Fabric.
I selected the “Industry Solutions” experience in the bottom-left hand corner and selected ‘Sustainability solution’. Just like I had done before to deploy the ESG data estate capability.
However, this time around I selected Microsoft Azure emissions insights capability.
After selecting Microsoft Azure emissions insights, a screen appears explaining the purpose of the solution.
In addition, it shows a diagram on the right-hand side that shows that you can consume your Azure emissions data into one Lakehouse and then extract the data to another Lakehouse and perform aggregations on it for reporting purposes.
As you can see, just like with the ESG data estate capability all the items contain the solution name as a prefix. With this in mind, when I reference items in in the rest of this post, I will refer to their prefix as ‘{Prefix}’. For easier reading.
I clicked the ‘Deploy to workspace’ button on this screen to deploy the capability.
Creating demo data in Microsoft Azure emissions insights
Once the capability was deployed, I looked into creating demo data. So, I opened up the ‘{Prefix}_LoadEmissionsDataToTables_INTB’ notebook (the name will vary depending on the name of your solution).
This notebook first declares some parameters before dropping existing tables. It then extracts the contents of a parquet file in the ‘{Prefix}_IngestedRawData_LH’ Lakehouse and loads them into a table in the ‘(Prefix)_AggregatedData_LH’ Lakehouse.
I checked all the cells were populated correctly before clicking the ‘Run all’ button for the notebook. Once completed I confirmed that the resource table was populated in the ‘(Prefix)_AggregatedData_LH’ Lakehouse.
I then opened the ‘{Prefix}_GenerateAggregateTables_INTB’ notebook. Which essentially inserts aggregated values from the resource table into a new table. Plus, it adds a new column afterwards.
I checked the cells looked okay before selecting the ‘Run all’ button. Afterwards, I confirmed that a new emissions_summary table was populated in the ‘(Prefix)_AggregatedData_LH’ Lakehouse.
Sample report
After creating all the demo tables I then decided to test the sample emissions report that comes with the capability.
In order to do this, I first had to configure the ‘{Prefix}_AnalyzeAzureEmissionsDataset_DTST’ semantic model that comes with the capability.
Afterwards, I was able to open up the sample AnalyzeAzureEmissions report that comes with the capability.
You can compare either subscriptions or resources in the default report as well. Which can bring an extra element to discussions about deployment choices.
Of course, you can also look to customize this report or create your own reports instead.
You can read about setting up this report in more detail in the ‘Visualize data and analytics‘ documentation online.
Viewing your own metrics
Working with demo data is all well and good but what if you want to view your own Microsoft Azure emission metrics? Well, you can look to consume your own data into the raw Lakehouse by using the Cloud for Sustainability API. Which is currently in preview.
From there, you can change the ‘RAW_DATA_EMISSIONS_ABFSS_PATH’ parameter in the ‘{Prefix}_LoadEmissionsDataToTables_INTB’ notebook to point to the location of your own data instead.
One key point I want to highlight about this method is that by default this notebook drops the existing tables. Which you can change if required.
If you just want to view your Azure emissions data instead of consuming it in Microsoft Fabric you have alternative options.
For example, you can look to use the Azure carbon optimization service. For a more lightweight offering you can also look to use the Emissions Impact Dashboard for Azure in Microsoft Fabric.
To manage expectations, both of the above offerings only work with certain Azure account types. For instance, the Cloud and Sustainability API documentation shows a list of supported accounts. In addition, it must be configured by somebody who has the relevant permissions.
Final words about the Microsoft Azure emissions insights capability
I hope that me sharing the results of my initial testing of the Microsoft Azure emissions insights capability within Microsoft Fabric inspires has raised awareness about this offering.
Because I find it really interesting that we can analyze the emissions of our Azure usage.
However, I must admit that I would like this to be extended to other Azure account types. So that more people can view their emissions insights.
Of course, if you have any comments or queries about this post feel free to reach out to me.
Be First to Comment