System and method to uniformly manage operational life cycles and service levels
A system and a method to manage a data center, the method including, for example, retrieving a physical topology of a service; determining from the physical topology a concrete type of a resource for the service; and selecting an actual instance of the resource in the data center. The actual instance having the concrete type and the actual instance selected such that a consumption of the actual instance does not violate at least one of a constraint and a policy.
1. A system comprising:
at least one processor; and
a non-transitory computer-readable medium including instructions, including of a microkernel, which, in response to being executed by the at least one processor, are configured to cause the system to perform operations comprising:
deploy a plurality of service managers for the management services based on stored profiles;
manage life cycle operations of the plurality of service managers and provide lookup responses to identify services, the plurality of service managers configured to manage life cycle operations for a plurality of management services and a plurality of services independent of the microkernel, the managing of the life cycle operations of the plurality of service managers including restarting a given service manager of the plurality of service managers based on a detection that a performance of the given service manager is within a threshold of not satisfying a service-level objective;
recall respective stored topologies for each of the plurality of management services to facilitate initialization of the life cycle operations of the management services; and
pass the respective stored topologies to associated ones of the plurality of service managers for deploying the plurality of management services as part of the initialization.
2. The system of claim 1 , wherein the plurality of management services include at least one of a resource manager, a provisioning manager, a dispatcher, a lock manager, and a configuration manager.
3. The system of claim 1 , further comprises an event system configured to transport the service-level objective from managed resources to subscribed controllers included in the plurality of service managers.
4. The system of claim 1 , wherein the detection that the performance of the given service manager is within the threshold of not satisfying the service-level objective is based on probability derived from an analysis of the service-level objective over time.
5. The system of claim 4 , wherein the service-level objective includes achievement of a key performance indicator, the key performance indicator including at least one of a response time or a throughput.
6. The system of claim 4 , further comprising a remedying engine configured to decrease the probability by reallocating resources associated with the plurality of services at runtime.
7. The system of claim 6 , wherein the remedying engine is further configured to perform the reallocating of the resources associated with the plurality of services at runtime such that a cost of the resources associated with the plurality of services at runtime is balanced with a benefit of satisfying the service-level objective.
8. The system of claim 6 , wherein the reallocating is based on an accessing of a plurality of physical topologies corresponding to the plurality of service managers that are stored in a configuration repository.
9. A method performed by at least one processor executing a microkernel, the method comprising:
deploying a plurality of service managers for the management services based on stored profiles;
managing life cycle operations of the plurality of service managers and providing lookup responses to identify services, using a plurality of service managers to manage life cycle operations for a plurality of management services and a plurality of services, the managing of the life cycle operations of the plurality of service managers including restarting a given service manager of the plurality of service managers based on a detection that a performance of the given service manager is within a threshold of not satisfying a service-level objective;
recalling respective stored topologies for each of the plurality of management services to facilitate initialization of the life cycle operations of the management services; and
passing the respective stored topologies to associated ones of the plurality of service managers for deploying the plurality of management services as part of the initialization.
10. The method of claim 9 , wherein the plurality of management services include at least one of a resource manager, a provisioning manager, a dispatcher, a lock manager, and a configuration manager.
11. The method of claim 9 , further comprising transporting the service-level objective from managed resources to subscribed controllers included in the plurality of service managers.
12. The method of claim 9 , wherein the detection that the performance of the given service manager is within the threshold of not satisfying the service-level objective is based on probability derived from an analysis of the service-level objective over time.
13. The method of claim 12 , wherein the service-level objective includes achievement of a key performance indicator, the key performance indicator including at least one of a response time or a throughput.
14. The method of claim 12 , further comprising decreasing the probability by reallocating resources associated with the plurality of services at runtime.
15. The method of claim 14 , further comprising performing the reallocating of the resources associated with the plurality of services at runtime such that a cost of the resources associated with the plurality of services at runtime is balanced with a benefit of satisfying the service-level objective.
16. The method of claim 15 , wherein the reallocating is based on an accessing of a plurality of physical topologies corresponding to the plurality of service managers that are stored in a configuration repository.
17. A non-transitory machine readable storage medium embodying a set of instructions, including of a microkernel, in which the instructions, in response to being executed by at least one processor, cause a system to perform operations, the operations comprising:
deploying a plurality of service managers for the management services based on stored profiles;
managing life cycle operations of the plurality of service managers and providing lookup responses to identify services that are using the plurality of service managers to manage life cycle operations for a plurality of management services and a plurality of services, the managing of the life cycle operations of the plurality of service managers including restarting a given service manager of the plurality of service managers based on a detection that a performance of the given service manager is within a threshold of not satisfying a service-level objective;
recalling respective stored topologies for each of the plurality of management services to facilitate initialization of the life cycle operations of the management services; and
passing the respective stored topologies to associated ones of the plurality of service managers for deploying the plurality of management services as part of the initialization.
18. The non-transitory machine readable storage medium of claim 17 , wherein the plurality of management services include at least one of a resource manager, a provisioning manager, a dispatcher, a lock manager, and a configuration manager.
19. The non-transitory machine readable storage medium of claim 17 , the operations further comprising transporting the service-level objective from managed resources to subscribed controllers included in the plurality of service managers.
20. The non-transitory machine readable storage medium of claim 17 , wherein the detection that the performance of the given service manager is within the threshold of not satisfying the service-level objective is based on probability derived from an analysis of the service-level objective over time.