Hey there James Tordoff
Thats a great question and thanks for using QandA platform,
lemme try claring this up for you.
Provisioned
This option is like reserving a set number of lanes on the highway just for yourself. You pay a fixed amount for a certain number of request units per second (RU/s), regardless of whether you're using them or not.
It's great for predictable workloads where you know exactly how much traffic you'll have.
If you can accurately estimate your peak retrieval times and are okay with paying for those reserved lanes even when you're not using them, this could be a good choice.
Serverless
Think of this as a pay-as-you-go model. You only pay for what you use, kind of like paying for tolls based on the distance you travel.
It's perfect for unpredictable workloads because you're not locked into any fixed capacity. Plus, you don't have to worry about over-provisioning or under-provisioning.
In your case, since you only have occasional retrieval, serverless could save you money during quieter periods, as you're not paying for resources you're not using.
https://learn.microsoft.com/en-us/search/?terms=%20Serverless%20Cosmos&category=Documentation
If this helps kindly accept the answer thanks much.