Systems and methods of maintaining availability of requested network resources

• US 7,590,746 B2
• Filed: 06/09/2003
• Issued: 09/15/2009
• Est. Priority Date: 06/07/2002
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Sparkes, Andrew Michael, Watkins, Mark Robert, Slater, Alastair Michael
• Primary Examiner(s): Walsh; John B.
• Application Number:
  US10/457,003
• Publication Number:
  US 20040047354A1
• Time in Patent Office: 2,290 Days
• Field of Search: None
• US Class Current: 709/229
• CPC Class Codes: H04L 41/0896 Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests

Abstract

The availability of requested network resources in a network of servers, each having associated network-servable resources of a specific kind is maintained. Resource data and resource-serving application software serve out the resource data and dynamically distribute capacity to serve out servable resources between the resource servers dependent on demand for the resources. The capability of the network as a whole to serve out a particular resource dependent upon the level of demand for the particular resource is thereby varied. The capacity to serve out servable resources is distributed between the resources without necessarily moving the resource data itself between the servers by distributing: resource-serving application software and/or installation software adapted to install resource-serving application software.

Claims

1. A method of maintaining availability of requested network resources of a network of servers each having associated network-servable resources of a specific kind, the network-servable resources including resource data and resource-serving application software adapted to serve out said resource data, the method comprising:

dynamically distributing capacity to serve out servable resources between said resource servers dependent upon demand for said resources, thereby varying capability of said network as a whole to serve out a particular resource dependent upon the level of demand for said particular resource,

wherein

said distributing of capacity to serve out servable resources comprises distributing at least one of;

(i) resource-serving application software necessary to serve out the particular resource; and

(ii) installation software adapted to install said resource-serving application software;

to

at least one server selected to serve out said particular resource without moving the resource data itself between servers; and

said at least one server is selected based on at least one of

(a) whether said at least one server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

(b) projected access requirements based upon previous access demands placed upon said at least one server.

2. A method according to claim 1 wherein an additional particular resource, including both resource data and resource-serving application software, is created on said network so as to increase capacity of said network to serve out said particular resource.

3. A method according to claim 2 wherein resource-serving application software of a specific kind is copied from a first resource server to a second resource server, thereby increasing the capacity of said network to serve out resource data in a format or protocol associated with said specific kind of resource-serving application protocol.

4. A method according to claim 3 further comprising at a different time freeing capacity in said network to serve out other said resources by reducing serving capacity of a different resource.

5. A method according to claim 3 comprising distributing said resources, or at least said resource data, between said resource servers in a homogenous manner so that each resource server has substantially the same resources, or resource data of said specific kind.

6. A method according to claim 3 comprising asymmetrically, non-homogeneously, distributing said resources of said specific kind so that not all resource servers having resources of said specific kind have the same resources of said specific kind available for serving.

7. A method according to claim 1 wherein an additional resource server, not part of said network at a first time, joins said network in order to increase capacity of said network to serve out said particular resource,

8. A method according to claim 7 wherein said additional resource server has resource data comprising at least one of said particular resource and resource-serving application software adapted to serve out said resource data copied to it to configure said additional resource server to serve out said particular resource.

9. A method according to claim 1 wherein said resource servers have port addresses and a particular resource servable from a first port address is made available for serving from a second port address by allocating appropriate resource-serving application software in said second port address.

10. A method of maintaining availability of requested network resources of a network of servers each having associated network-servable resources of a specific kind, the network-servable resources including resource data and resource-serving application software adapted to serve out said resource data, the method comprising:

dynamically distributing capacity to serve out servable resources between said resource servers dependent upon demand for said resources, thereby varying capability of said network as a whole to serve out a particular resource dependent upon the level of demand for said particular resource,

wherein

said distributing of capacity to serve out servable resources comprises distributing at least one of;

(i) resource-serving application software;

(ii) installation software adapted to install resource-serving application software;

to at least one server selected to serve out said particular resource without necessarily moving the resource data itself between servers;

said at least one server is selected based on

(a) the bandwidth available to said at least one server,

(b) whether said at least one server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

(c) projected access requirements based upon previous access demands placed upon said at least one server; and

said resource servers have port addresses and a particular resource servable from a first port address is made available for serving from a second port address by allocating appropriate resource-serving application software to said second port address.

11. A method according to claim 10 wherein said first and second port addresses are associated with the same server.

12. A method according to claim 10 wherein said first and second port addresses are associated with different resource servers.

13. A method according to claim 10 which comprises automatically dynamically expanding capacity of a video serving network to serve out a specific video, comprising said network servable resource, by either of (i) copying said specific video to another said resource server, or (ii) by having videos servable out from port addresses of said servers and by making said specific video available for serving out from an increased number of port addresses, or by making said specific video available for serving out from a port address with an increased capacity to serve out said specific video.

14. A method according to claim 10 wherein said resources or said data content comprise one of:

rich media;

video;

audio.

15. A network comprising:

resource servers each having a deliverable resource of a specific kind available for serving out, and

a resource manager adapted to receive a demand indicator input indicative of demand for a particular one of said deliverable resources of said specific kind and to output a resource-varying output adapted to vary capability of said network as a whole to serve out said particular one resource, said resource-varying output being adapted to vary the level available for use of resource-serving application software running, installed, or installable from application-installation software on said resource servers of said network;

wherein said resource-varying output is generated based on

(a) the bandwidth available to each said server

(b) whether each said server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

c) projected access requirements based upon previous access demands placed upon each said server.

16. A network according to claim 15 wherein said resource-varying output comprises a resource migration signal for causing either:

(i) said particular one resource to migrate or be copied from one resource server, or (ii) said resource-serving application software, or installation software for said resource serving application software, to migrate or be copied from one server to another.

17. A network according to claim 15 wherein said resource-varying output is adapted to cause copying of said particular one resource, or of said application software, or of said installation software, so as to cause said particular one resource to be at least one of present arid servable out on said one server and also on said another server.

18. A network according to claim 17 wherein said resource-varying output is adapted to increase capacity of said network as a whole to serve out said particular one resource.

19. A network according to claim 15 wherein said resource servers have port addresses and wherein said particular one resource is available for serving from one poll address of a said server and wherein said resource varying output causes a second port address of a said server to have said particular resource available for serving.

20. A network according to claim 19 wherein said second port address is a port address of the same resource server that has said first port address.

21. A network according to claim 19 wherein said second port address is provided on a different said resource server to that said resource server which has said first port address.

22. A network according to claim 15 wherein said resource-varying output is adapted to reduce serving capacity of said network as a whole to serve out a selected resource.

23. A network according to claim 15 wherein said resource-varying output causes said resources of said specific kind to be distributed homogeneously over said resource servers that are in said network.

24. A network according to claim 15 wherein said resource-varying output causes said resources of said specific kind to be distributed asymmetrically, non-homogeneously, over said resource servers so that at least one of said resource servers has different resources of said specific kind available for serving out.

25. A network according to claim 15 wherein said resource-varying output is adapted to cause an additional server, external to said network, to be contacted and joined with said network.

26. A network according to claim 25 wherein said resource-varying output also causes a selected resource, or resource-serving application software, or installation software for resource serving application software, present on said network to be copied to said additional server and made available for use on said additional server.

27. A network according to claim 15 wherein said resource manager comprises a said resource server.

28. A network according to claim 27 wherein each said resource server is configured as said resource manager.

29. A network comprising;

plurality of resource servers adapted to serve out servable recourses of a specific kind;

a usage monitor; and

a resource manager;

said usage monitor being adapted to monitor usage of said servable resources and to determine if a usage level of a particular said resource has changed significantly and to communicate such changes in usage level to said resource manager,

said resource manager being adapted to take into account

(i) the bandwidth available to each said server,

(ii) whether each said server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

(iii) projected access requirements based upon previous access demands placed upon each said server,

and to do at least one of

(a) introduce an additional server to said network, or start tip a resource serving application on a networked resource server if usage of said servable resources has increased significantly, and

(b) remove said specific resource from a networked resource server, or shut down a resource-serving application on a networked resource server if said usage of said servable resources has decreased.

30. A resource management server for use with a plurality of resource servers having resources of the same particular kind available for serving out, said resource management server being operable to evaluate demand for a particular one of said resources of said specific kind, in use, in order to vary capacity of said network to serve out said particular one resource in response to changes in demand for said particular one resource, and

wherein said resource management server is operable to take into account

(i) the bandwidth available to each said server,

(ii) whether each said server is currently being accessed either or serving out the same particular resource or for serving out another resource, and

(iii) projected access requirements based upon previous access demands placed upon each said server,

and to alter the number of resource-serving applications of a specific kind adapted to serve out resource in a specific particular format or protocol that is running, installed, or in installation software format, available on said resource servers as a whole and available for future use.

31. A resource management server to claim 30 being further operable to cause said particular one resource, or application software, to be copied from a first server in said network to a second server in said network so as to increase capacity of said network to serve out said particular one resource.

32. A resource management server according to claim 30 being further operable to cause initiation of serving out of said particular one resource from a port address of one of said resource servers by bringing up running appropriate resource-serving application software, said port address being either (a) on a resource server that is already serving out said particular one resource from a different port address, or (b) on a resource server that was not previously serving out said particular one resource.

33. A resource management server according to claim 30 having resource of said specific kind and which comprises a resource server operable to serve out resource of said specific kind.

34. A content management server for use with a content server network, said content management server comprising

a servable content monitor adapted to monitor levels of usage of servable content of a specific kind, and

a content availability manager adapted to change the availability of specific servable content depending upon demand for said specific servable content;

said content management server being operable to take into account

(i) the bandwidth available to said server network,

(ii) whether said server network is currently being accessed either for serving out the same specific content or for serving out another content; and

(iii) projected access requirements based upon previous access demands placed upon said server network,

and to perform at least one of

(a) the function of an additional resource server to a network, or starting up a content serving application on a content server if levels of usage of a particular content have increased significantly, and

(b) removing content from a networked content server, or shutting down a content serving application on a content server if levels of usage of a particular content have decreased significantly.

35. A content management server according to claim 34 which is itself a content server configured for serving out content of said specific kind.

36. A resource management server for use with a plurality of networked resource servers having resources of a common particular kind available for serving out, said resource management server comprising

a demand evaluator adapted to evaluate demand for a particular one of said servable out resources of said specific kind based on,

(i) the bandwidth available to each said server,

(ii) whether each said server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

(iii) projected access requirements based upon previous access demands placed upon each said server, and

a capacity modifier adapted to modify capacity of said network to serve out said particular one resource, said capacity modifier being arranged to be responsive to evaluated demand signals from said demand evaluator and being adapted to distribute resource-serving application software over servers of said network so as to maintain unused available capacity to serve out resource.

37. A method of maintaining availability of requested network resources in a network of servers each having associated network-servable resources of a specific kind and port addresses, the method comprising;

dynamically distributing servable resources between said resource servers dependent upon demand for said resources, thereby varying capability of said network as a whole to serve out a particular resource dependent upon the level of demand for said particular resource, and

making a particular resource servable from a first port address available for serving from a second port address associated with the same server as said first port address, said server being selected for serving out the particular resource based on

(a) the bandwidth available to said server,

(b) whether said server is currently being accessed either for serving out the same particular resource or for serving out another resource, and

(c) projected access requirements based upon previous access demands placed upon said server.