Reserved concurrency and throttling

Reserved concurrency sets a hard cap on how many instances of a Lambda function can run simultaneously. When you set reserved concurrency to 5, Lambda guarantees those 5 slots are always available for your function — but it also means a 6th simultaneous request will be throttled.

This is useful when your function writes to a downstream resource (like a database) that can only handle a limited number of connections. Without reserved concurrency, a sudden spike in traffic could overwhelm the database with hundreds of parallel Lambda executions.

There's a trade-off: reserved slots are subtracted from your account's total concurrency pool (default 1,000). Setting reserved concurrency to 100 on one function leaves only 900 for all other functions in the account.

What happens when a function with reserved concurrency of 5 receives a 6th simultaneous request?