Dynamic system level agreement provisioning

Patent metadata

Field	Value
Publication number	US-10764159-B2
Application number	US-201414496868-A
Country	US
Kind code	B2
Filing date	Sep 25, 2014
Priority date	Dec 4, 2013
Publication date	Sep 1, 2020
Grant date	Sep 1, 2020

Abstract

According to one aspect of the present disclosure a method and technique for dynamic system level agreement provisioning is disclosed. The method includes: identifying, by a data processing system of a computing environment service provider, service level criteria for a customer of computing services; determining characteristics of the computing environment; identifying a time period for providing the computing services; evaluating one or more utility functions defining service level variables; and automatically determining, by the data processing system, a service level agreement (SLA) provision for the customer based on the one or more utility functions.

First claim

What is claimed is:

1. A method, comprising:
identifying, by a data processing system of a computing environment service provider, service level criteria for requested computing services;
determining characteristics of the computing environment;
identifying a time period for providing the computing services;
constructing, by the data processing system, one or more utility functions defining service level variables corresponding to the service level criteria for the time period, the utility functions including a weight applied to each service level variable, the weight representing a probability of a violation of the respective service level variable;
evaluating the one or more utility functions against a plurality of computing resource configurations allocatable from a resource pool of the computing environment; and
allocating a select one of the computing resource configurations from the resource pool of the computing environment, by the data processing system, for the requested computing services based on the one or more utility functions being maximized based on the plurality of computing resource configurations without exceeding a limit utility value.

2. The method of claim 1 , further comprising:
monitoring the computing environment over the time period; and
automatically updating the one or more utility functions based on a change to the computing environment.

3. The method of claim 1 , wherein identifying service level criteria for the customer comprises receiving a specification of a response time parameter, an availability parameter, and a capacity parameter.

4. The method of claim 1 , further comprising selecting the weight based on historical data of the computing environment.

5. The method of claim 1 , further comprising selecting the weight based on a most probable cause of a respective service level variable violation for a respective computing resource configuration.

6. A method, comprising:
receiving, by a data processing system of a computing environment service provider, service level criteria for requested computing services;
determining characteristics of the computing environment;
identifying a time period for providing the computing services;
constructing, by the data processing system, for each of a plurality of different computing resource configurations a combined utility function defining service level variables corresponding to the service level criteria for the time period, the combined utility function assessing the joint effects of the service level variables;
selecting a weight to apply to each service level variable in the respective combined utility functions, the weight representing a probability of a violation of the respective service level variable;
evaluating the combined utility function for each of the plurality of different computing resource configurations; and
allocating, by the data processing system, a select one of the plurality of different resource configurations for the requested computing services based on a maximum value of the evaluated combined utility functions.

7. The method of claim 6 , further comprising evaluating changes in the weights occurring over the time period.

8. The method of claim 6 , further comprising selecting the weights based on historical data of the computing environment.

9. The method of claim 6 , further comprising selecting the weights based on a most probable cause of a respective service level variable violation for a respective computing resource configuration.

Assignees

• IBM

Inventors

• Chakra Al
• Ghosh Rahul
• Mcmahan Rocky D
• Raghavan Vasanth

Classifications

• H04L41/5012

  determining service availability, e.g. which services are available at a certain point in time · CPC title

• H04L41/5025

  by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade · CPC title

• H04L41/5006 Primary

  Creating or negotiating SLA contracts, guarantees or penalties · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title