A Lambda function calls a third-party API that enforces a requests-per-minute limit. If requests exceed that limit, the API returns rate-limit errors. How should the developer configure the Lambda function to avoid hitting the third-party rate limits?

Setting the Lambda function's reserved concurrency directly controls the maximum number of concurrent invocations for that specific function. By matching this limit to the third-party API's requests-per-minute limit, you prevent your Lambda function from overwhelming the API and incurring rate-limit errors. Reserved concurrency ensures that your function will not exceed this specified number of concurrent executions, regardless of overall account concurrency. Reducing memory allocation won't affect the rate of requests to the third-party API. Provisioned concurrency is for pre-initializing execution environments to reduce cold start latency, not for limiting concurrent requests to external services. Increasing the Lambda function timeout only allows individual invocations more time to complete, but does not limit the number of concurrent invocations.