The Yield After Minutes field helps you prevent any processor from being monopolized by a long-running map or reduce job.
Here’s how this setting works: During the map and reduce stages, after each function invocation, the system checks to see how long the map or reduce job has been running. If the amount of elapsed time has surpassed the number of minutes identified in the Yield After Minutes field, the job gracefully ends its execution, and a new job is created to take its place. The new job has the same priority as the old one, but a later timestamp. This behavior is known as yielding.
By default, Yield After Minutes is set to 60, but you can enter any number from 3 to 60.
The system never interrupts a function invocation for this limit. Also, the system never ends a map or reduce job before the limit has been reached, but only after it has been passed. For that reason, the degree to which the limit is surpassed varies depending on the duration of your function invocation. For example, if the Yield After Minutes limit is 3 minutes, but your function takes 15 minutes to complete, then in practice the job yields after 15 minutes, not 3 minutes.
Yielding is also affected by a governance limit. This limit is 10,000 usage units for each map and reduce job. This limit works in the same way as the Yield After Minutes limit: The system waits until after each function invocation ends to determine whether the usage-unit limit has been surpassed. If it has, the job yields, even if the Yield After Minutes limit has not been exceeded.