Container-as-a-service (CaaS) controller for monitoring clusters and implemeting autoscaling policies
Embodiments described herein are generally directed to a controller of a managed container service that facilitates autoscaling based on bare metal machines available within a private cloud. According to an example, a CaaS controller of a managed container service monitors a metric of a cluster deployed on behalf of a customer within a container orchestration system. Responsive to a scaling event being identified for the cluster based on the monitoring and an autoscaling policy associated with the cluster, a BMaaS provider associated with the private cloud may be caused to create an inventory of bare-metal machines available within the private cloud. Finally, a bare metal machine is identified to be added to the cluster by selecting among the bare-metal machines based on the autoscaling policy, the inventory and a best fit algorithm configured in accordance with a policy established by or on behalf of the customer.
1. A system comprising:
a processing resource; and
a non-transitory computer-readable medium, coupled to the processing resource, having stored therein instructions that when executed by the processing resource cause the processing resource to:
monitor a metric associated with operation of a cluster deployed on behalf of a customer of a managed container service within a container orchestration system;
responsive to a scaling event being identified for the cluster based on the monitoring and an autoscaling policy associated with the cluster, cause a Bare-Metal-as-a-Service (BMaaS) provider associated with the private cloud to create an inventory of a plurality of bare-metal machines available within the private cloud; and
identify a bare metal machine to be added to the cluster by selecting among the plurality of bare-metal machines based on the auto scaling policy, the inventory and a best fit algorithm configured in accordance with a policy established by or on behalf of the customer.
2. The system of claim 1 , wherein the instructions further cause the processing resource to receive the autoscaling policy as part of a definition of the cluster from a user of the customer, wherein the definition of the cluster includes information regarding a machine configuration desired to be used by the cluster.
3. The system of claim 1 , wherein the scaling event comprises triggering of a scale out action of a rule of the autoscaling policy.
4. The system of claim 1 , wherein the instructions further cause the processing resource to provide the customer with flexible capacity by using cloud-bursting or workload shifting when demand associated with the cluster exceeds a capacity of the private cloud.
5. The system of claim 1 , wherein the policy expresses a goal of minimizing excess resources of the plurality of bare metal machines, wherein the inventory includes information that quantifies a value of each resource for each of the plurality of bare-metal machines, and wherein the instructions further cause the processor to:
identify a subset of the plurality of bare-metal machines having a type and a quantity of resources that satisfy a machine specification identified by the autoscaling policy;
for each machine in the subset, computing an excess resource metric based on an amount of resources of the machine that are in excess of resources required to satisfy the machine specification; and
selecting as the particular bare metal machine a machine of the subset having the excess resource metric indicative of a least amount of excess resources.
6. The system of claim 1 , wherein the instructions further cause the processing resource to request the BMaaS provider to create the inventory via a BMaaS portal associated with the BMaaS provider.
7. The system of claim 6 , wherein deployment of the cluster was responsive to a request received via a Container-as-a-Service (CaaS) portal.
8. The system of claim 7 , wherein the CaaS portal and the BMaaS portal are operable within a public cloud.
9. The system of claim 1 , wherein the system comprises a CaaS controller operable within the public cloud.
10. The system of claim 1 , wherein the system comprises a CaaS controller operable within the private cloud.
11. A non-transitory machine readable medium storing instructions that when executed by a processing resource of a computer system cause the processing resource to:
monitor a metric associated with operation of a cluster deployed on behalf of a customer of a managed container service within a container orchestration system;
responsive to a scaling event being identified for the cluster based on the monitoring and an autoscaling policy associated with the cluster, cause a Bare-Metal-as-a-Service (BMaaS) provider associated with the private cloud to create an inventory of a plurality of bare-metal machines available within the private cloud; and
identify a bare metal machine to be added to the cluster by selecting among the plurality of bare-metal machines based on the auto scaling policy, the inventory and a best fit algorithm configured in accordance with a policy established by or on behalf of the customer.
12. The non-transitory machine readable medium of claim 11 , wherein the instructions further cause the processing resource to receive the autoscaling policy as part of a definition of the cluster from a user of the customer, wherein the definition of the cluster includes information regarding a machine configuration desired to be used by the cluster.
13. The non-transitory machine readable medium of claim 11 , wherein the scaling event comprises triggering of a scale out action of a rule of the autoscaling policy.
14. The non-transitory machine readable medium of claim 13 , wherein the instructions further cause the processing resource to provide the customer with flexible capacity by using cloud-bursting or workload shifting when demand associated with the cluster exceeds a capacity of the private cloud.
15. The non-transitory machine readable medium of claim 11 , wherein the policy expresses a goal of minimizing excess resources of the plurality of bare metal machines, wherein the inventory includes information that quantifies a value of each resource for each of the plurality of bare-metal machines, and wherein the instructions further cause the processor to:
identify a subset of the plurality of bare-metal machines having a type and a quantity of resources that satisfy a machine specification identified by the autoscaling policy;
for each machine in the subset, computing an excess resource metric based on an amount of resources of the machine that are in excess of resources required to satisfy the machine specification; and
selecting as the particular bare metal machine a machine of the subset having the excess resource metric indicative of a least amount of excess resources.
16. The non-transitory machine readable medium of claim 11 , wherein the instructions further cause the processing resource to request the BMaaS provider to create the inventory via a BMaaS portal associated with the BMaaS provider.
17. A method comprising:
monitoring, by a processing resource of a Container-as-a-Service (CaaS) controller of a managed container service, a metric associated with operation of a cluster deployed on behalf of a customer of the managed container service within a container orchestration system;
responsive to a scaling event being identified for the cluster based on the monitoring and an autoscaling policy associated with the cluster, causing, by the processing resource, a Bare-Metal-as-a-Service (BMaaS) provider associated with the private cloud to create an inventory of a plurality of bare-metal machines available within the private cloud; and
identifying, by the processing resource, a bare metal machine to be added to the cluster by selecting among the plurality of bare-metal machines based on the autoscaling policy, the inventory and a best fit algorithm configured in accordance with a policy established by or on behalf of the customer.
18. The method of claim 17 , further comprising receiving, by the processing resource, the autoscaling policy as part of a definition of the cluster from a user of the customer, wherein the definition of the cluster includes information regarding a machine configuration desired to be used by the cluster.
19. The method of claim 17 , wherein the scaling event comprises triggering of a scale out action of a rule of the autoscaling policy.
20. The method of claim 17 , further comprising causing, by the processing resource, the customer to be provided with flexible capacity by using cloud-bursting or workload shifting when demand associated with the cluster exceeds a capacity of the private cloud.