Changing Serverless Compute Resources

Scenarios

Cloud Eye monitors CPU usage and memory usage of serverless instances. When certain conditions are met, Cloud Eye automatically scales up or down the serverless compute resources.

Constraints

When a compute change is triggered, if resources are insufficient, a node with desired specifications is created on a physical machine with sufficient resources. If the resources on the primary node are insufficient, there will be a primary/standby switchover.

Scaling up Compute

Scale-up is triggered when either of the following conditions is met:

• The CPU usage is greater than 90% for 5 seconds and the interval between two scale-up operations is at least 5 seconds.
• The memory usage is greater than 90% for 5 seconds and the interval between two scale-up operations is at least 5 seconds.

Scaling down Compute

Scale-down is triggered when either of the following conditions is met:

• The CPU usage is less than 30% for 15 seconds and the interval between two scale-down operations is at least 15 seconds.
• The memory usage is less than 30% for 15 seconds, the CPU usage is less than 70% for 15 seconds, and the interval between two scale-down operations is at least 15 seconds.