Performance optimization through run-time quality governance

摘要

The present invention provides a method and system for optimizing the performance of a service according to the load on the service. The invention includes altering the quality of various sub-ordinate services that are a part of the service and in turn achieving systemic service performance levels that satisfy prescribed service level agreements.

主权项

1. A method for managing performance of a server, the server receiving a plurality of requests from one or more users and generating an output for each request received by using one or more resources, the method comprising:
dynamically monitoring instantaneous load on the server at predefined intervals of time; continuously tracking a set of performance parameters of the server, the set of performance parameters comprising:
(i) weights assigned to one or more sub-ordinate servers and resources upon which the server inherently depends for collecting and collating intermediary results to generate the output, and(ii) ratio of response time of one sub-ordinate server to response time of another sub-ordinate server of the one or more subordinate servers;identifying configurable rules that governs quality of the output generated for the request received; and
modifying the configurable rules to alter quality of the output based on the instantaneous load and the set of performance parameters, the quality of output being altered by lowering the quality of the output to within acceptable limits as defined in a set of base SLAs (Service Level Agreements) and without adding more resources, wherein the modification of the configurable rules includes: modifying a mapping mechanism to identify and map a relevant result upon identifying any undesirable result being generated by the server in response to the request received, and modifying the weights assigned to one or more sub-ordinate servers and resources.