×

Enterprise data-driven system for predictive resource provisioning in cloud environments

  • US 9,606,840 B2
  • Filed: 06/27/2013
  • Issued: 03/28/2017
  • Est. Priority Date: 06/27/2013
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method for predicting resource consumption in cloud infrastructures, the method being executed using one or more processors and comprising:

  • receiving, by the one or more processors, event data from one or more enterprise data sources, the event data comprising an interaction between an enterprise and a cloud infrastructure;

    determining, by the one or more processors, that an event associated with the event data is a known event by matching the event with an existing event entry in a matching table using a matching format comprising at least one of a class of the event, an average memory allocation consumed, and an average processing units consumed;

    retrieving, by the one or more processors, resource consumption information associated with the event; and

    providing, by the one or more processors, a resource consumption schedule to the cloud infrastructure, the resource consumption schedule indicating discrete values of a future demand using different levels of probability of the future demand, the future demand being associated with resources expected to be consumed during execution of the event that are quantified based on a stepwise refinement of a time, at which or during which the event is to be executed.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×