System and method for managing the performance of an enterprise application
AssigneeInfosys Limited
InventorsSANDEEP BHAGAT, VENKATA REDDY DONTHIREDDY
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to a framework to improve the predictability of performance problems in an enterprise application. In one embodiment, the present disclosure produces accurate predictive analysis by taking into consideration all the factors including workload, historical performance, environments, configurations, and data volumes etc. which are significant for determining the future performance characteristics of the system. In addition, the present disclosure calculates the actual impact of the determined performance characteristics on the business.
Claims8
A system for managing the performance of an enterprise application, said system comprising:
- a central server 200 configured to collect workload and performance metrics in a performance message protocol from a plurality of servers 210;
- a workload modeler 202 configured to create a current workload model using the workload metrics collected over a predefined period of time;
- a trend analyzer 204 configured to produce a forecast workload model using the workload model created over a predefined period of time;
- an enterprise infrastructure modeler 224 configured to track and maintain configuration changes in the plurality of servers;
- a predictive analyzer 206 configured to receive a plurality of inputs to predict a future workload for managing the enterprise application performance, wherein the plurality of inputs comprises the current workload model, the forecast workload model, current infrastructure deployment model, the configuration changes in the plurality of servers and external factors including data volume growth and application configuration changes;
- and
- a business analytics unit 208 configured to analyze effect of the future workload on business performance.
A method for managing the performance of an enterprise application, said method comprising the steps of:
- collecting workload metrics in performance message protocol from a plurality of servers;
- creating a current workload model using the workload metrics collected over a predefined period of time;
- producing a forecast workload model using the workload model created over a predefined period of time;
- tracking and maintaining configuration changes in the plurality of servers;
- receiving a plurality of inputs to predict a future workload for managing the enterprise application performance,
- wherein the plurality of inputs comprises the current workload model, the forecast workload model, current infrastructure model, the configuration changes in the plurality of servers and external factors including data volume growth and application configuration changes;
- and
- analyzing effect of the future workload on business performance.
Dependencies are inferred from the claim text.