API Gateway throttling in AWS - Step-by-Step Execution

Process Flow - API Gateway throttling

Client sends API request

↓

API Gateway receives request

↓

Check current request rate

↓

Rate < Limit

↓

Forward request

↓

Response sent

API Gateway checks the number of requests per second. If below limit, it forwards the request. If limit is reached, it rejects with a throttle error.

Execution Sample

AWS

Client sends 5 requests per second
API Gateway limit is 3 requests per second
Requests 1,2,3 pass
Requests 4,5 get throttled

Shows how API Gateway allows requests up to the limit and throttles the rest.

Process Table

Request Number	Current Rate (req/sec)	Condition (Rate < Limit)	Action	Response
1	1	True	Forward request	Success
2	2	True	Forward request	Success
3	3	True	Forward request	Success
4	4	False	Reject request	Throttle Error
5	4	False	Reject request	Throttle Error
-	-	-	-	Requests 4 and 5 rejected because rate limit 3 req/sec exceeded

💡 Requests 4 and 5 exceed the limit of 3 requests per second, so they are throttled.

Status Tracker

Variable	Start	After Request 1	After Request 2	After Request 3	After Request 4	After Request 5
Current Rate (req/sec)	0	1	2	3	4	4
Action Taken	None	Forward	Forward	Forward	Reject	Reject
Response	None	Success	Success	Success	Throttle Error	Throttle Error

Key Moments - 2 Insights

Why are requests 4 and 5 rejected even though they are sent by the client?

Because the current request rate (4 req/sec) exceeds the API Gateway limit (3 req/sec), so throttling rejects these requests as shown in execution_table rows 4 and 5.

Does the API Gateway reset the request count immediately after throttling?

No, the request count is tracked per second window. Requests 4 and 5 are still counted in the same second, so they remain throttled until the next window.

Visual Quiz - 3 Questions

Look at the execution table, what is the response for request number 3?

A Throttle Error

B Success

C Timeout

D No Response

💡 Hint

Check the 'Response' column for request number 3 in the execution_table.

At which request number does the condition 'Rate < Limit' become false?

A Request 2

B Request 3

C Request 4

D Request 5

💡 Hint

Look at the 'Condition (Rate < Limit)' column in the execution_table.

If the API Gateway limit was increased to 5 requests per second, what would happen to request 5?

A It would be forwarded successfully

B It would be throttled

C It would timeout

D It would be ignored

💡 Hint

Compare the current rate and limit in the variable_tracker and execution_table.

Concept Snapshot

API Gateway throttling limits how many requests pass per second.
If requests exceed the limit, extra requests get rejected with a throttle error.
This protects backend services from overload.
Limits are set per API stage or method.
Throttling resets every second window.

Full Transcript

API Gateway throttling controls the number of requests allowed per second. When a client sends requests, API Gateway counts how many requests arrive in the current second. If the count is below the set limit, the request is forwarded to the backend and a success response is sent. If the count reaches or exceeds the limit, further requests are rejected with a throttle error. This prevents backend overload. For example, if the limit is 3 requests per second and the client sends 5 requests, the first 3 succeed and the last 2 are throttled. The request count resets every second window.