Self-regulated resource management of distributed computer resources

• US 6,516,350 B1
• Filed: 06/17/1999
• Issued: 02/04/2003
• Est. Priority Date: 06/17/1999
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Wood, Stephen P., Lumelsky, Leon L., Manohar, Nelson R.
• Primary Examiner(s): Najjar, Saleh
• Application Number:
  US09/335,261
• Time in Patent Office: 1,328 Days
• Field of Search: 709/217, 709/219, 709/220, 709/221, 709/222, 709/223, 709/225, 709/226, 709/229, 709/231, 709/235, 710/36-40, 710/107, 710/200, 710/240-244, 711/150, 711/151
• US Class Current: 709/226
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A distributed system for the management of distributed resources interconnected by a computer network and comprised of a plurality of limited supply resources (such as those associated with multimedia content servers, e.g., bandwidth, CPU, storage, etc.,). The system comprises a plurality of server resources, preferably with target multimedia content, real time processing engines, etc., and a plurality of intermediary system resources (such as directories and resource monitors). The resources in any particular server are classified or partitioned as global or local, where the ratio of global to local is specified and controlled by the system. The system assigns clients across local and global resources and coordinates the placement of replicas of target content across global resources. The placement is dynamic and performed when necessary based on the analysis of utilization patterns of target content and replicas by pluralities of clients.

Claims

1. A system for regulating server resources in a distributed wide area network comprising servers interconnected by a computer networking environment, said resources including multimedia object resources and resources for storing and streaming said multimedia objects to requesting clients, the system comprising:

first service management means implementing objects for receiving client requests for streaming object resources from a server to said client, generating statistics associated with demand for said object resources, and automatically mapping said requests to available servers in said network based on predetermined criteria, including user profiles for characterizing client resources and user preferences relating to receipt of said object resource, and per-flow services relating to one or more selected from the group comprising, transcoding object content, offered pricing for said content, billing, authorizing and authenticating users, and encrypting/decrypting data content, and wherein said first management means further includes one or more first negotiator objects for locating a service object in said distributed network capable of handling said request according to said predetermined criteria, generating and placing inquiries related to said received requests, and negotiating requests with a service object; and

,

second system management means implementing objects for automatically shaping capacity of object resources at said servers in said distributed network based on said demand statistics, current available capacity of said servers and a willingness factor associated with each server for providing said requested resource, said second management including means for automatically replicating an object associated with said objects resources and temporarily placing said object replicas at one or more server devices having available capacity according to said predetermined criteria, wherein performance for placing requests of said first management means has an input of and considers said capacity shaping provided by said second management means.

2. The system as claimed in claim 1, wherein said client resources includes one or more of processing power, memory, an available bandwidth for receiving said object content at a client resource.

3. The system as claimed in claim 1, wherein said user preferences includes preferable compression of said content, cost limitations for said content, and interactivity level including real-time interaction with flow of said object resource content.

4. The system as claimed in claim 1, wherein said criteria includes willingness of said servers to provide said requested services, and available capacity of said server for providing a streaming resource.

5. The system as claimed in claim 1, further including facade means for interfacing per-flow services with said first management means, said facade means enabling dynamic insertion of both resources and content in a manner transparent to clients.

10. The system as claimed in claim 1, wherein said servers in said network are normalized as comprising a storage bin of a size for storing object resource content, said second management means managing placement of content across storage bins so that access optimizations may be made.

11. The system as claimed in claim 10, wherein said servers in said network are normalized by representing a server as comprising an associated service bin characterized as a tuple comprising memory, CPU allocation and streaming bandwidth required for provisioning services.

12. The system as claimed in claim 10, wherein a object resource streaming connection between a server and a client is instantiated by binding a storage bin of an associated server to its service bin while taking into account one or more quality of service parameters.

13. The system as claimed in claim 12, wherein a streaming connection is instantiated by binding a storage bin of an associated server to its service bin while taking into account one or more quality of service parameters.

14. The system as claimed in claim 10, wherein said second management means manages placement of content across storage bins according to geography of demand.

15. The system as claimed in claim 14, wherein said second management means manages placement of content across storage bins in a manner so as to support peak demand.

16. The system as claimed in claim 14, further including a signaling protocol clients and content servers enabling control servers to monitor the placement of requests from clients onto said servers.

17. The system as claimed in claim 1, wherein the wide area network comprises the Internet.

6. A system for regulating server resources in a distributed wide area network comprising servers interconnected by a computer networking environment, said resources including multimedia object resources and resources for storing and streaming said multimedia objects to requesting clients, the system comprising:

first service management means implementing objects for receiving client requests for streaming object resources from a server to said client, generating statistics associated with demand for said object resources, and automatically mapping said requests to available servers in said network based on predetermined criteria, wherein said first management means further includes one or more first negotiator objects for locating a service object in said distributed network capable of handling said request according to said predetermined criteria, generating and placing inquiries related to said received requests, and negotiating requests with a service object, wherein first negotiator objects invoke a first directory service means for locating said service object, said first directory service means comprising a mapping of service objects to locations of associated object replicas at said servers, and said first negotiator objects invoke a second directory service for mapping available distributed servers with an available capacity rating, wherein said second directory service further maps a willingness indicator to an associated server, said willingness indicator indicating degree of willingness of a server to receive placement inquiries;

second system management means implementing objects for automatically shaping capacity of object resources at said servers in said distributed network based on said demand statistics, current available capacity of said servers and a willingness factor associated with each server for proving said requested resource, said second management including means for automatically replicating an object associated with said objects resources and temporarily placing said object replicas at one or more server devices having available capacity according to said predetermined criteria, wherein performance for placing requests of said first management means has an input of and considers said capacity shaping provided by said second management means wherein said second management means includes one or more service broker objects representing available servers in said network, said service broker objects interacting with said second negotiator objects for placing generated object replicas associated with an object resource to said global servers throughout said distributed network.

7. The system as claimed in claim 6, wherein said second management means includes means for generating object replicas associated with object resources for placement at global servers throughout said distributed network, said second management means further including one or more second negotiator objects for distributing object replicas over available servers.

8. The system as claimed in claim 7, wherein said second management means generates object replicas on demand.

9. The system as claimed in claim 7, wherein a number of generated object replicas associated with an object resource is determined in accordance with predicted demand for a given object.

18. A method for regulating server resources in a distributed wide area network comprising servers interconnected by a computer networking environment, said resources including multimedia object resources and resources for storing and streaming said multimedia objects to requesting clients, the method comprising:

providing objects for receiving client requests for streaming object resources from a server to said client, generating statistics associated with demand for said object resources, and automatically mapping said requests to available servers in said network based on predetermined criteria, wherein said step of automatically mapping requests further includes steps of,

implementing one or more first negotiator objects for locating a service object in said distributed network capable of handling said request according to said predetermined criteria,

generating and placing inquiries related to said received requests, and

negotiating requests with a service object, which step of negotiating requests includes the step of negotiating for per-flow services and implementing one or more facades for dynamically inserting optional including both resources and content in a manner transparent to clients, said per-flow services including transcoding object content, offered pricing for said content, billing, authorizing and authenticating users, and encrypting/decrypting data content;

automatically shaping capacity of object resources at said servers in said distributed network based on said demand statistics, current available capacity of said servers and a willingness factor associated with each server capable of providing said requested resource;

automatically replicating an object associated with said objects resources and temporarily placing said object replicas at one or more server devices having available capacity according to said predetermined criteria, wherein performance for placing requests for streaming object resources has an input of and considers said capacity shaping step; and

characterizing client resources according to user profiles and user preferences relating to receipt of said object resource, and implementing said user profiles and user preferences as said predetermined criteria.

19. The method as claimed in claim 18, wherein said step of locating a service object includes referencing a first directory service for mapping service objects to locations of associated object replicas at said servers.

20. The method as claimed in claim 19, further including referencing a second directory service for mapping available distributed servers with an available capacity rating.

21. The method as claimed in claim 20, further including referencing a second directory service for mapping available distributed servers with a willingness indicator indicating a degree of willingness of a server to receive placement inquiries.

22. The method as claimed in claim 18, wherein said step of automatically replicating an object associated with said objects resources and temporarily placing said object replicas at one or more server devices includes implementing or more second negotiator objects for distributing object replicas over available servers.

23. The method as claimed in claim 18, wherein said step of generating statistics associated with demand for said object resources includes:

predicting a demand for a given object resource; and

,

determining a number of generated object replicas associated with said predicted demand.

24. The method as claimed in claim 22, further including providing one or more service broker objects representing available servers in said network for interaction with said second negotiator objects to place generated object replicas associated with an object resource to said global servers throughout said distributed network.

25. The method as claimed in claim 18, further including characterizing said servers in said network as comprising a storage bin of a size for storing object resource content, and managing placement of content across storage bins.

26. The method as claimed in claim 25, further including characterizing said servers in said network as comprising an associated service bin including a tuple comprising memory, CPU allocation and streaming bandwidth required for provisioning services.

27. The method as claimed in claim 26, further including the step of instantiating an object resource streaming connection between a server and a client by binding a storage bin of an associated server to its service bin while taking into account one or more quality of service parameters.

28. The method as claims in claim 18, further including implementing a signaling protocol between clients and content servers for monitoring the placement of requests from clients onto said servers.

29. The method as claimed in claim 18, wherein the wide area network comprises the Internet.