Latency sensitive
Event driven workloads
~5-10 second startup
Long running compute jobs greater than 15 minutes
More than 10GB of memory required
Container as a Service (CaaS)
Charges you for the vCPU and memory resources of your containerized applications use per second.
Charges you per invocation and duration of each invocation based on the memory configured
Environment remains in a warm state
Function as a Service (FaaS)
Short running compute jobs less than 15 minutes
Least Operational Burden, Least Flexibility
More Operational Burden, More Flexibility