Methods and systems for adaptive capacity management

摘要

Techniques to adaptively manage service requests within a multi-server system. In one embodiment, a service request and a service rule associated with the service request are received. Data about operating parameters of at least one server in a multi-server system are also received as part of a feedback loop. A response to the service request based on the service rule and the operating parameters is determined. Execution of the service request may be modified according to a tiered service rule based on the at least one server reaching a capacity threshold. The modification includes omitting an action in execution of the service request.

主权项

1. A computer implemented method, comprising:
implementing an application service proxy, by a computer system, for a multi-server system to receive a service request from an application service executing within the multi-server system, wherein the service request is an inter-service communication within the multi-server system; receiving, by the computer system, a tiered service rule associated with the received service request; monitoring, by the computer system, operating parameters of at least one server in the multi-server system, wherein the operating parameters include load placed by each of one or more application services executed on each of the at least one server in the multi-server system; and modifying, by the computer system, execution of the service request to be performed at the at least one server based on a tier-specific threshold level of the tiered service rule that is satisfied by the monitored operating parameters, wherein modifying the execution of the service request includes narrowing service scope of the service request by omitting an action among a plurality of computer-implemented actions requested in the service request; determining, by the computer system, a response to the service request based on the tiered service rule and the monitored operating parameters; and configuring, based on the service request, the multi-server system to consume at least a server resource of the multi-server system to execute the plurality of computer-implemented actions minus the omitted action, wherein the server resource includes processor execution cycles, memory space provided by a memory device, network bandwidth provided by a network adapter, or any combination thereof.