API Gateway throttling in AWS - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding API Gateway Throttling Limits

What is the default rate limit for AWS API Gateway throttling per account per region?

A 100,000 requests per second

B 1,000 requests per second

C 10,000 requests per second

D 5,000 requests per second

💡 Hint

Think about the default soft limits AWS sets for API Gateway to protect backend services.

AWS API Gateway has a default regional throttling limit of 5,000 requests per second per account. This helps prevent backend overload.

❓ Architecture

intermediate

Designing Throttling for Multiple API Stages

You have multiple stages (dev, test, prod) in your API Gateway. How can you apply throttling limits independently for each stage?

A Apply throttling only at the account level, stages share the same limits

B Use Lambda authorizers to enforce throttling per stage

C Set usage plans with API keys scoped to each stage

D Create separate API Gateway accounts for each stage

💡 Hint

Think about how usage plans and API keys can control access and limits per stage.

Usage plans with API keys allow you to set throttling and quota limits per stage by associating keys with specific stages.

❓ scaling

advanced

Handling Bursty Traffic with API Gateway Throttling

Your API experiences sudden bursts of traffic exceeding the throttling limits. Which approach best helps to handle this without dropping requests?

A Enable burst capacity and use usage plans with quotas

B Increase the throttling limits manually without any other changes

C Disable throttling to allow all requests through

D Use CloudFront in front of API Gateway to cache responses

💡 Hint

Consider AWS features that allow temporary traffic spikes without rejecting requests.

API Gateway supports burst capacity to handle sudden traffic spikes temporarily. Usage plans with quotas help manage overall usage.

❓ tradeoff

advanced

Tradeoffs of Client-Side vs Server-Side Throttling

Which is a key tradeoff when choosing client-side throttling over server-side throttling in API Gateway?

A Client-side throttling guarantees no requests are dropped by the server

B Client-side throttling reduces server load but relies on client compliance

C Server-side throttling requires clients to implement retry logic

D Server-side throttling cannot be configured in API Gateway

💡 Hint

Think about control and reliability differences between client and server throttling.

Client-side throttling reduces load on the server but depends on clients to respect limits, which may not always happen.

❓ estimation

expert

Estimating Maximum Requests with Throttling and Burst Capacity

Your API Gateway has a steady rate limit of 5,000 requests per second and a burst capacity of 10,000 requests. How many requests can it handle in 2 seconds during a sudden traffic spike?

A 20,000 requests

B 15,000 requests

C 10,000 requests

D 5,000 requests

💡 Hint

Consider how burst capacity adds to the steady rate limit over time.

In the first second, API Gateway can handle up to 15,000 requests (10,000 burst + 5,000 steady), and 5,000 steady requests in the second second, totaling 20,000 requests over 2 seconds.