Gen2 Capacity Monitoring Memory in Metrics

Currently, the fact that there is no insight into memory allocation during refresh in Gen2 metric apps, this is costing us $1500/mo in required tier capacity upgrade to handle our data refresh! So yeah, this seems like a pretty big deal! I'd say +1000 on the votes for this "idea" especially since this was a feature in the previous Gen!

As of 25/11/2022 I can confirm the Memory Metric has stopped working for people that migrated to gen2 earlier than required. For us the Gen1 metric has been working for a full year prior to the migration, and the idea that "the systems archetecture has changed" seems implausable.

If the archetecture has indeed changed, and the ability to use memory as a metric to regulate the service is no longer an option, selling PBI embedded based on memory should also be removed. Why should I pay for 5gb of memory, if nothing you can provide me gives evidence that this is required or being utilitised?

The metric Item size in capacity metrics will show peak memory usage for workloads that emit it to us. It is an optional column in the Fabric capacity metrics app.

Metrics app scope is currently focused on capacity utilization metrics, for detailed workload analytics i.e. detailed analysis of AS you may want to look into Log Analytics.

Look for Tracing with Log Analytics and Power BI in youtube by Guy in a Cube (can't link directly currently). The Governance team is also working on a platform monitoring solution for Fabric that will be targeted for workload specific analytics, details on this will be discussed at the upcoming Fabric conference and semester release notes for Fabric.

Please provide Gen 2 metrics also in Azure metrics Explorer. it is very essential.