Elasticity engine for availability management framework (AMF)

A system adapted to dynamically respond to a change in a workload managed by Availability Management Framework (AMF), wherein the workload is represented by a set of service instances (SIs) assigned to service units (SUs) of a first service group (SG) hosted on nodes of a cluster, wherein the first SG uses a redundancy model to protect the set of Sis and shares one or more nodes with one or more other SGs, the system comprising:

• memory to store a configuration;
• and
• one or more processors coupled to the memory, the one or more processors adapted to:
• receive a notification of the change in the workload indicating the type of workload change and the impacted SI in the set of SIs;
• select, based on the type of workload change and the redundancy model used by the first SG, a strategy from a collection of SG-level strategies and cluster-level strategies to accommodate the change in the workload by adjusting an amount of resources available to at least the impacted SI, wherein each SG-level strategy changes first one or more configuration attributes within the first SG according to the redundancy model used by the first SG, and wherein each cluster-level strategy changes second one or more configuration attributes within one or more of the other according to respective redundancy models used by the other SGs within the cluster, wherein the select comprises select from the cluster-level strategies in response to determining that none of the SG-level strategies can accommodate the change in the workload;
• and
• cause the AMF to apply configuration changes to configuration attributes in accordance with the selected strategy to respond to the change in the workload.