Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 511 discussion

Option C suggests using Amazon Aurora On-Demand PostgreSQL-Compatible databases for each development environment. This option provides the benefits of Amazon Aurora, which is a high-performance and scalable database engine, while allowing you to pay for usage on an on-demand basis. Amazon Aurora On-Demand instances are typically more cost-effective for individual development environments compared to the provisioned capacity options.

Guys, when you use the pricing calculator the cost between option B and C is really close. I doubt anyone wants to test on your knowledge of exact pricings in your region. I think that "On Demand" being explicitly specified in option C and not being specified in option B is the main difference here the exam wants to test. In that case I'd assume that option B means a constantly running instance and not "On Demand" which would make the choice pretty obvious. Again, I don't think AWS exam will test you on knowing that a single AZ is cheaper by 0,005 cents than Aurora :D

B RDS single az is the cheapest. Aurora needs at least two azs.

I choose C because Aurora on-Demand is Aurora Serverless: The Aurora Serverless is cost effective. Scale out fine-grained increments to provide just the right number of database resources and pay only for capacity consumed.

Single-AZ DB instances cheaper

Single AZ more cost effective

1 instance(s) x 0.245 USD hourly x (4 / 24 hours in a day) x 730 hours in a month = 29.8083 USD ---> Amazon RDS PostgreSQL instances cost (monthly) 1 instance(s) x 0.26 USD hourly x (4 / 24 hours in a day) x 730 hours in a month = 31.6333 USD ---> Amazon Aurora PostgreSQL-Compatible DB instances cost (monthly)

C is correct because B is cheaper but they don't mention to stop the DB when not in use

On-Demand is cheaper that Aurora or RDS because of low weekly usage

We have environments that are used on average 4 hours per workday = 20 hours per week. So with option C (Aurora on-demand aka serverless) we pay for 20 hours per week. With option B (RDS) we pay for 168 hours per week (the answer does not mention anything about automating shutdown etc.). So even if Aurora Serverless is slightly more expensive than RDS, C is cheaper because we pay only 20 (not 168) hours per week.

Aurora on demand is (a little) more expensive than Aurora Aurora is more expensive than RDS single instance So cost effectiveness == RDS. (B)

AWS Services Calculator is showing B cheaper by less than a dollar for the same settings for both. I used "db.r6g.large" for RDS (Single-AZ) and Aurora and put 4 hours/day.

Amazon RDS Single AZ is cheaper than Aurora Multi-AZ

Aurora instances will cost you ~20% more than RDS MySQL Given the running hours the same. Also Aurora is HA.

… just trying to trick you. Aurora on demand is Aurora Serverless.

Aurora allows you to pay for the hours used. 4 hour every day, you only need 1/6 cost of 24 hours per day. You can check the Aurora pricing calculator.

The key factors: RDS Single-AZ instances only run the DB instance when in use, minimizing costs for dev environments not used full-time RDS charges by the hour for DB instance hours used, versus Aurora clusters that have hourly uptime charges PostgreSQL is natively supported by RDS so no compatibility issues S3 Object Select (Option D) does not provide full database functionality Aurora (Options A and C) has higher minimum costs than RDS even when not fully utilized