Container-as-a-service (CAAS) controller for private cloud container cluster management

A method comprising:

• receiving, by a Container-as-a-Service (CaaS) controller, cluster information associated with a request to create a container cluster on behalf of a customer, the cluster information comprising master and worker node sizes for the container cluster and a version of the container cluster;
• receiving, by the CaaS controller, an inventory of bare metal machines available within a private cloud of the customer;
• identifying, by the CaaS controller, a particular bare metal machine for the container cluster based on the inventory of the bare metal machines, the cluster information, and a best fit algorithm, the best fit algorithm configured in accordance with a policy established by the customer, wherein identifying the particular bare metal machine for the container cluster includes:
• creating a set of candidate machines based on the inventory of the bare metal machines, the set of candidate machines comprising a subset of the bare metal machines that have sufficient resources to satisfy the request;
• calculating an excess resource metric for each candidate machine in the set of candidate machines;
• and
• selecting, as the particular bare metal machine, a particular candidate machine in the set of candidate machines having the excess resource metric indicative of a least amount of excess resources;
• and
• creating, by the CaaS controller, the container cluster on the particular bare metal machine.

A non-transitory machine readable medium storing instructions that when executed by a processing resource, cause the processing resource to:

• receive cluster information associated with a request to create a container cluster on behalf of a customer, the cluster information comprising master and worker node sizes for the container cluster;
• receive an inventory of bare metal machines available within a computing cloud;
• receive a learned historical pattern associated with a user of the customer;
• identify a particular bare metal machine for the container cluster based on the inventory of the bare metal machines, the cluster information, the learned historical pattern, and a best fit algorithm, the best fit algorithm configured in accordance with a policy established by the customer, wherein the particular bare metal machine is identified by:
• creating a set of candidate machines based on the inventory of the bare metal machines, the set of candidate machines comprising a subset of the bare metal machines that have sufficient resources to satisfy the request;
• calculating an excess resource metric for each candidate machine in the set of candidate machines;
• and
• selecting, as the particular bare metal machine, a particular candidate machine in the set of candidate machines having the excess resource metric indicative of a least amount of excess resources;
• and
• create the container cluster on the particular bare metal machine.