Optimizing advanced cluster nodes
You can optimize master and worker nodes by provisioning enough resources for each node type.
Optimizing master nodes
Provision a master node with enough resources to manage the worker nodes in the advanced cluster.
The following table lists the master node configuration based on the number of worker nodes:
Worker node count | Master node configuration |
|---|
1-10 | 4 CPUs and 8 GB memory |
11-100 | 8 CPUs and 32 GB memory |
Optimizing worker nodes
Provision worker nodes with enough resources to process jobs.
The following table lists the worker node configuration based on the advanced cluster type:
Advanced cluster type | Worker node configuration |
|---|
Fully-managed cluster | Minimum of 8 CPUs and 32 GB memory. For better performance, use 16 CPUs and 64 GB memory. |
Local cluster | Minimum of 4 CPUs and 16 GB memory. For better performance, use 8 CPUs and 32 GB memory. |
The number of worker nodes that you provision depends on your SLA.