Hierarchical resource management for a computing utility

• US 8,655,997 B2
• Filed: 01/30/2004
• Issued: 02/18/2014
• Est. Priority Date: 01/30/2004
• Status: Active Grant

• Current Assignee: ServiceNow Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Eilam, Tamar, Hunt, Guerney D. H., Miller, Sandra D., Mummert, Lily B.
• Primary Examiner(s): TOWFIGHI, AFSHAWN M
• Application Number:
  US10/587,618
• Publication Number:
  US 20080216082A1
• Time in Patent Office: 3,672 Days
• Field of Search: 709217-229
• US Class Current: 709/223
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/503 Resource availability
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

This invention provides for the hierarchical provisioning and management of a computing infrastructure which is used to provide computing services to the customers of the service provider that operates the infrastructure. Infrastructure resources can include those acquired from other service providers. The invention provides architecture for hierarchical management of computing infrastructures. It allows the dynamic provisioning and assignment of resources to computing environments. Customers can have multiple computing environments within their domain. The service provider shares its resources across multiple customer domains and arbitrates on the use of resources between and within domains. The invention enables resources to be dedicated to a specific customer domain or to a specific computing environment. Customers can specify acquisition and distribution policy which controls their use of resources within their domains.

Claims

1. A method comprising providing an automatic hierarchical management of a computing infrastructure for at least one domain for an entity, said step of providing hierarchical management comprising:

obtaining a hierarchical representation of said at least one domain, wherein the at least one domain comprises a plurality of sub-domains and said hierarchical representation including;

a list of computing environments to be managed, at least one policy controlling acquisition of at least one resource from resource libraries for said at least one domain, and the plurality of sub-domains the at least one policy for controlling the number of reserve resources available to process requests from the at least one domain and the sub-domains, the resources and reserve resources comprising a reserve pool of resources pre-allocated for handling requests for the at least one sub-domain and a free pool of resources for handling requests for any of the plurality of sub-domains;

receiving at least one request for resources from said at least one sub-domain, wherein the at least one request includes a request for an amount of resources that exceeds an available amount of resources in the reserve pool and the free pool; and

accepting the at least one request based on an expectation that additional resources will be acquired in time for allocation, wherein at least one of the allocated resources is not available at a time of the reservation is made.

2. A method as recited in claim 1, further comprising:

deriving a set of resources required for said list of computing environments in constructing said hierarchical management, and

providing resources for said set of resources to said at least one domain.

3. A method as recited in claim 1, further comprising:

further comprising updating said at least one policy of the representation;

further comprising utilizing library services;

further comprising associating each computing environment with a particular sub-domain;

wherein the step of utilizing includes reserving a set of resources required by said list of computing environments;

further comprising acquiring the set of resources and using at least one resource from said set of resources;

wherein said at least one domain is a plurality of domains;

wherein at least one domain from said at least one domain is a sub domain of another domain;

further comprising associating at least one library service from said library services with at least one collector;

wherein both the quantity and types of base resources change over time, wherein said method is employed in providing service on-demand;

wherein said at least one resource is a base or composite resource;

further comprising organizing said at least one resource into a service offered to a plurality of customers;

further comprising allocating base resources to a library service;

further comprising formulating composite resources from base resources satisfying a service description;

further comprising allocating composite resources to a library service; and

further comprising allocating services to a library service.

4. A method as recited in claim 3, wherein at least one of said at least one domain is a root domain.

5. A method as recited in claim 4, further comprising a requesting computing environment making a request for a particular combination of resources checking said representation of the acquisition policy of said requesting computing environment to verify that satisfaction of the request for the particular of resources is within the acquisition policy of said requesting computing environment;

and

repeating the step of checking for all parent collector of said requesting collector until any root collector is reached.

6. A method as recited in claim 5, further comprising determining if the acquisition policy is satisfied all the way to any root collector;

if the acquisition policy is satisfied the request is granted otherwise the request is denied.

7. A method as recited in claim 4, further comprising making a request for a particular combination of resources;

determining a starting collector to start a search for the combination of resources;

checking if the starting collector has at least one resource from said combination of resources, said at least one resource being a located resource;

checking if there is at least one library which includes at least one resource from said combination of resources, said at least one resource being a located resource;

repeating the step of checking at each collector from a starting collector to any root collector;

if all resources of said combination are located resources reserving all located resources, otherwise denying the request.

8. A method as recited in claim 7, further comprising calling arbitration to continue locating all resources from said combination of resources.

9. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for providing hierarchical management of at least one domain for a computing utility, said method steps comprising the steps of claim 1.

10. An article of manufacture comprising a non-transitory computer readable storage medium tangibly storing instructions, which when executed by a computer implement provisioning of hierarchical management of at least one domain for a computing utility, the computer readable program code means in said article of manufacture comprising computer readable program code means for causing a computer to effect the steps of claim 1.

11. An apparatus comprising means for providing hierarchical management of at least one domain for a computing utility, said means for providing hierarchical management comprising:

means for obtaining a hierarchical representation of said at least one domain, wherein the at least one domain comprises a plurality of sub-domains and said representation including;

a list of computing environments to be managed, at least one policy controlling acquisition of at least one resource from composite resources for said at least one domain, and of the plurality of sub-domains the at least one policy for controlling the number of reserve resources available to process requests from the at least one domain and the sub-domains, the resources and reserve resources comprising a reserve pool of resources pre-allocated for handling said predicted at least one request for the at least one sub-domain and a free pool of resources for handling a request for any of the plurality of sub-domains;

means for receiving at least one request for resources from said at least one sub-domain, wherein the at least one request includes a request for an amount of resources that exceeds an available amount of resources in the reserve pool and the free pool; and

accepting the at least one request based on an expectation that additional resources will be acquired in time for allocation, wherein at least one of the allocated resource is not available at a time of the reservation is made.

12. An architecture for a computing utility comprising an apparatus to provide at least one service for a plurality of clients, said apparatus comprising:

a Base Resource Distribution Service configured to allocate a plurality of resources to said at least one service, the plurality of resources comprising a free resource pool comprising one or more resources for handling requests for any of a plurality of clients and a reserve resource pool comprising one or more resources pre-allocated for handling requests for a first client;

said Base Resource Distribution Service having at least one collector;

at least one Provisioned and Managed Resource Service coupled to said Base Resource Distribution Service to provision and manage said resources for said at least one service; and

at least one Base Resource Library Service coupled to said Base Resource Distribution Service to provide reservation and allocation of resources, said Base Resource Distribution Service configured to provide a reservation of resources in excess of currently available resources in the computing utility based on an expectation that additional resources will be added to the computing utility in time for allocation.

13. An architecture as recited in claim 12, wherein said Base Resource Distribution Service comprises:

at least one collector, each collector anchoring a representation of a particular domain and holds polices of said particular domain and holds resources reserved for said particular domain; and

an arbiter coupled to said base resource distribution service and available to each of said collectors, said arbiter provides dynamic resource allocation to each collector of said computing utility.

14. An architecture as recited in claim 12, wherein the architecture is used by an on-demand service.