Huge Spikes in Fabric Capacity Usage

[bw_chec]

Hi all,

I am using a F8 capacity.

On several occasions the capacity has randomly started failing to create clusters in notebooks due to too many API requests.

The workload didn't increase normally (i am the only engineer working in Fabric), running the same pipelines and notebook that had run previously in the day.

Looking at the capacity metrics app, I have some abnormally huge spikes in CU% over the past couple of weeks.

These spikes go up to 150000% CU? 

Any help appreciated

[v-jingzhan-msft]

Hi @bw_chec 

It's hard to tell what's going on from the current chart. You can try to select a peak point and click the Explore button at the bottom right to see which background jobs were running at that time and some other information of these jobs. 

Best Regards,
Jing
If this post helps, please Accept it as Solution to help other members find it. Appreciate your Kudos!