Required. Specifies the machine types, the number of replicas
for workers and parameter servers.
Unspecified Scale Tier.
A single worker instance. This tier is
suitable for learning how to use Cloud ML, and
for experimenting with new models using small
Many workers and a few parameter servers.
A large number of workers with many parameter
A single worker instance with a K80 GPU.
A single worker instance with a Cloud TPU.
The CUSTOM tier is not a set tier, but rather enables you to
use your own cluster specification. When you use this tier,
set values to configure your processing cluster according to
- You *must* set `ExecutionTemplate.masterType` to
specify the type of machine to use for your master node.
This is the only required setting.