Microsoft Azure Data FundamentalsDescribe how to work with relational data on AzureHard
A company is considering migrating its on-premises PostgreSQL database to Azure. The database experiences highly fluctuating workloads, with periods of intense activity followed by long periods of inactivity. The company wants to minimize costs by only paying for compute resources when the database is actively used. Which Azure Database for PostgreSQL deployment option best meets these requirements?
- ASingle Server (General Purpose)
- BFlexible Server (Burstable)
- CFlexible Server (Serverless)
- DHyperscale (Citus)
Show answer & explanationAnswer & explanation
Correct answer: C. Flexible Server (Serverless)
Azure Database for PostgreSQL - Flexible Server (Serverless compute tier) automatically scales compute capacity based on workload demand and automatically pauses when inactive, allowing the company to pay only for storage during idle periods, perfectly aligning with fluctuating workloads and cost minimization goals.
Why the other options are wrong
- A. Single Server (General Purpose) has provisioned compute that runs continuously, leading to higher costs during idle periods.
- B. Flexible Server (Burstable) is for low-cost development/test workloads or smaller applications with infrequent full capacity needs, but it doesn't automatically pause and resume like serverless.
- D. Hyperscale (Citus) is for horizontally scaling large analytical workloads across multiple nodes, not primarily for cost optimization of fluctuating transactional workloads through pausing.
Azure DB for PostgreSQL Serverless
A compute tier for Azure Database for PostgreSQL - Flexible Server that automatically scales compute capacity based on workload, and pauses when inactive, billing only for storage during idle periods.
- Optimizes costs for intermittent and unpredictable workloads.
- Automatically scales up and down compute resources.
- Automatically pauses during inactivity and resumes on connection.
Memory trick: Serverless: Pay for what you use, when you use.