Automatically identifying and right sizing instances

A non-transitory computer-readable memory (NTCRM) comprising instructions stored thereon that, in response to execution by a processor, are operable to cause the processor to:

• monitor a resource utilization of a set of resources of one or more instances, the resource utilization corresponding to a first level of performance and cost;
• determine, based on the resource utilization, whether there is an instance type for at least one of the one or more instances, with a resource profile, that will provide a second level of performance and cost that is closer to a default level of performance and cost than the first level of performance and cost;
• causing an evaluation of the recommended instance type when the recommended instance type is different from a current instance type of the at least one of the one or more instances, the evaluation includes provisioning, as part of an experiment, one or more instances of the recommended instance type to service work associated with an application cluster;
• and
• perform, based on the determination, performing one of making and not making a recommendation to replace the instance type of the at least one of the one or more instances.

A computer-implemented method for providing an instance advisory service for an Infrastructure as a Service (IaaS), comprising:

• monitoring a resource utilization of a set of resources of one or more instances, the resource utilization corresponding to a first level of performance and cost;
• determining, based on the resource utilization, whether there is an instance type for at least one of the one or more instances, with a resource profile, that will provide a second level of performance and cost that is closer to a default level of performance and cost than the first level of performance and cost;
• causing an evaluation of the recommended instance type when the recommended instance type is different from a current instance type of the at least one of the one or more instances, the evaluation includes provisioning, as part of an experiment, one or more instances of the recommended instance type to service work associated with an application cluster;
• and
• performing, based on the determining, one of making and not making a recommendation to replace the instance type of the at least one of the one or more instances.