DP-420 · Question #146
You are creating an Azure Cosmos DB for NoSQL account. You need to choose whether to configure capacity by using either provisioned throughput or serverless mode. Which two workload characteristics…
The correct answer is B. low average-to-peak traffic ratio E. intermittent traffic. Serverless mode is ideal for Azure Cosmos DB workloads with intermittent traffic and a low average-to-peak traffic ratio, as it charges only for consumed RUs.
Question
You are creating an Azure Cosmos DB for NoSQL account. You need to choose whether to configure capacity by using either provisioned throughput or serverless mode. Which two workload characteristics cause you to choose serverless mode? Each correct answer presents part of the solution. NOTE: Each correct selection is worth one point.
Options
- Ageo-distribution requirements
- Blow average-to-peak traffic ratio
- C25 TB of data in a single container
- Dhigh average-to-peak traffic ratio
- Eintermittent traffic
How the community answered
(36 responses)- A3% (1)
- B72% (26)
- C8% (3)
- D17% (6)
Why each option
Serverless mode is ideal for Azure Cosmos DB workloads with intermittent traffic and a low average-to-peak traffic ratio, as it charges only for consumed RUs.
Geo-distribution is a feature available with provisioned throughput mode, but not currently with serverless mode.
Serverless mode is cost-effective for workloads where the average throughput is significantly lower than the peak throughput, meaning requests are spiky rather than consistently high. This is precisely what a low average-to-peak traffic ratio indicates.
Serverless containers have a maximum storage limit of 1 TB, so 25 TB in a single container would exceed this limit and require provisioned throughput.
A high average-to-peak traffic ratio indicates a more consistent workload, which is often better suited for provisioned throughput to ensure stable performance and potentially lower cost at scale, as opposed to serverless.
Serverless mode automatically scales with intermittent traffic, meaning you only pay for the request units and storage consumed, avoiding the cost of provisioned throughput during idle periods. This characteristic perfectly aligns with the serverless billing model.
Concept tested: Cosmos DB serverless vs provisioned throughput
Source: https://learn.microsoft.com/azure/cosmos-db/serverless
Topics
Community Discussion
No community discussion yet for this question.