System and method for highly-scalable real-time and time-based data delivery using server clusters

• US 7,403,993 B2
• Filed: 07/24/2002
• Issued: 07/22/2008
• Est. Priority Date: 07/24/2002
• Status: Expired due to Fees

• Current Assignee: Espial Incorporated (Enghouse Systems Limited)
• Original Assignee: Kasenna, Inc. (Enghouse Systems Limited)
• Inventors: Thota, Laxmi, John, Ranjit, Wang, James, Muthukumarasamy, Jayakumar, Menon, Satish N., Horen, Robert
• Primary Examiner(s): Najjar; Saleh
• Assistant Examiner(s): Bruckart; Benjamin R
• Application Number:
  US10/205,476
• Publication Number:
  US 20040088412A1
• Time in Patent Office: 2,190 Days
• Field of Search: 709/231, 709/203, 709/226, 709/219, 709/273, 709/223, 709/244, 709/227, 709/232, 718/105, 725/115, 455/3.04
• US Class Current: 709/226
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: H04L 65/612 for unicast
• CPC Class Codes: H04L 65/80 Responding to QoS
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

The present invention provides loosely-coupled cluster systems comprising a plurality of servers based on storage attached to the plurality of servers. Videos, or other assets, are automatically replicated within the server system to increase the number of concurrent play requests serviceable. The server systems can detect spikes in demand that may exceed the guaranteed number of concurrent play requests serviceable and dynamically transfer the high-in-demand or ‘hot’ asset to servers in the cluster that do not have the video. Alternatively, instead of transferring the entire asset, varying length prefixes of the asset may be transferred depending on the availability of resources. The remainder of the asset is transferred in some embodiments on demand with sufficient buffering or other storage to guarantee playback to the user or subscriber according to the required quality of service (QOS).

Claims

1. A server system for time-based media streaming comprising:

a plurality of servers coupled for communication with each other, said plurality of servers including a first server; and

a computer readable storage medium at said first server storing therein asset information that is replicated in a computer readable storage medium associated with at least one other server;

wherein said first server is configured to;

receive a request for an asset from a client coupled to the server system;

determine if the request should be forwarded to another server before responding to the client, wherein said determination is based on existence of at least one of a plurality of conditions consisting of;

that the asset does not exist at the first server;

that sufficient resources do not exist to stream the asset from the first server; and

that a current load on the first server is over a threshold limit;

find in the server system a second server capable of streaming the asset in response to the determination that the request should be forwarded;

cause the request from the client to be sent to the second server after the second server is located; and

determine if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

2. A server system according to claim 1, wherein each server is further configured to periodically calculate at least one load factor corresponding to the server and broadcast the at least one load factor to other servers in the server system.

3. A server system according to claim 2, wherein each server is further configured to store load factors received from other servers, and wherein said second server is the least loaded among servers having the asset according to load factors stored at the first server.

4. A server system according to claim 2, wherein the at least one load factor corresponding to each server comprises an overall load factor.

5. A server system according to claim 4, wherein said overall load factor is computed based on information chosen from the group consisting of:

percent CPU used, memory used, network bandwidth available, and combinations thereof.

6. A server system according to claim 2 wherein each server in the server system stores a load update frequency indicating how often the server is to calculate the at least one load factor corresponding to the server and broadcast the at least one load factor to other servers in the server system.

7. A server system according to claim 1, further comprising a network switch coupled between the client and the plurality of servers.

8. A server system according to claim 1, wherein said asset comprises an audio or a video.

9. A server system according to claim 8, wherein said video comprises a movie video in MPEG-2 format for playback at 30 frames/second.

10. A server system according to claim 1, wherein each server is configured to broadcast an asset insert message to other servers in response to an asset being inserted in asset information stored at the server and to broadcast an asset delete message to other servers in response to an asset being removed from the asset information stored at the server.

11. A server system according to claim 1, wherein said asset information comprises metadata associated with an asset.

12. A server system according to claim 11, further comprising a shared storage system, wherein said shared storage system stores a plurality of media assets and each server comprises a computer storage device storing therein metadata associated with at least one of said plurality of media assets.

13. A server system according to claim 1, wherein each server in the server system stores a load threshold value.

14. A server system according to claim 1, wherein each server in the server system is further configured to communicate during an activation phase a list of assets stored at the server to other servers in the server system.

15. A server system according to claim 14, wherein each server having part of an asset stored at the server is configured to respond to a request for the asset by requesting transfer of the asset while streaming the part of the asset.

16. A server system according to claim 1, wherein said first server stores a hot asset count value and is further configured to keep track of received requests for assets and to replicate all or part of an asset to another server in response to a determination that a number of received requests for the asset exceeds the hot asset count value.

17. A server system according to claim 1, wherein said first server stores a hot asset count value and a hot asset period and is further configured to keep track of received requests for assets and to replicate part or all of the asset to another server in response to a determination that a number of received requests for the asset during a period equal in length to the hot asset period exceeds the hot asset count value.

18. A server system according to claim 1, further comprising a cluster management console configured to collect server information, asset information, load information and active stream counts associated with each server in the server system, the cluster management console being further configure to define the server system, add server to the server system, delete a server from the server system, and monitor asset requests and hot objects.

19. A server system according to claim 1, wherein said first server is further configured to cause the request from the client to be sent to the second server by informing the client to send the request to the second server.

20. A server system according to claim 1, wherein said time-based media streaming comprises real-time media streaming.

21. A server system according to claim 1, wherein said plurality of servers form a peer-to-peer server network.

22. A method for time-based streaming of assets, said method comprising:

receiving a request from a client for an asset at a first server;

determining, prior to streaming said asset, if said first server has said asset and if said first server has sufficient resources to stream said asset;

if said first server does not have said asset, or said first server does not have sufficient resources to stream said asset, attempting to identify a second server having said asset and sufficient resources to stream said asset;

informing the client to send the request directly to said identified second server prior to streaming said asset; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

23. A method for time-based streaming of assets and load-balancing, said method comprising:

receiving a request from a client for an asset at a first server having said asset and sufficient resources to stream said asset;

determining if said first server has a first server load level greater than a load threshold level; and

in response to the determination that said first server has a first server load level greater than a load threshold level;

attempting, prior to streaming said asset, to find a second server having said asset, sufficient resources to stream said asset, and a second server load level less than said first server load level;

informing the client to send the request to said second server if said second server is located;

or streaming said asset while maintaining a time-base for said streamed asset if said second server is not located

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

24. A method according to claim 23, wherein said load threshold level is predetermined and adjustable.

25. A method according to claim 23, further comprising periodically calculating said first server load level and broadcasting said first server load level to other servers including the second server.

26. A method for time-based streaming of assets performed by a first server in a server system comprising a plurality of servers, said method comprising:

receiving a request for an asset from a client;

determining prior to responding to the client if said first server has said asset and if said first server has sufficient resources to stream said asset;

if said first server has said asset and sufficient resources to stream said asset, determining if said first server has a load level less than a load threshold value; and

if said first server has a first server load level greater than a load threshold level, attempting, prior to streaming said asset, to locate a second server having said asset, sufficient resources to stream said asset, and a second server load level less than said first server load level; and

(i) informing the client to send the request to said second server if said second server is located;

or (ii) streaming said asset if said second server is not located; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

27. A method according to claim 26, wherein said load threshold level is predetermined and adjustable.

28. A method according to claim 27, further comprising periodically calculating said first server load level and broadcasting said first server load level to other servers including the second server.

29. A method according to claim 26, wherein said determining if said asset is hot comprises determining if a number of requests for said asset during a predetermined hot asset period exceeds a predetermined hot asset count.

30. A method according to claim 26, wherein said determining if said asset is hot comprises determining if a number of requests for said asset during a predetermined hot asset period equals or exceeds a predetermined hot asset count.

31. A method according to claim 26, wherein said copying information comprises copying metadata associated with said asset to said second server.

32. A method according to claim 26, wherein said detennining if said asset exists at said first sewer comprises determining if information about said asset is stored in a first computer readable storage medium associated with said first server.

33. A method according to claim 32, wherein said information about said asset comprises information selected from the group consisting of said asset, metadata associated with said asset, a prefix of said asset, and combinations thereof.

34. A method according to claim 26, wherein said first server has said asset if all or a portion of said asset exists at said first server, said method further comprising:

requesting transfer of said asset from another server to said first server.

35. A method according to claim 26, wherein a part of said asset exists at said first server, said method further comprising:

requesting transfer of said asset from a centralized storage system to said first server.

36. A computer program product for use in conjunction with a first server having at least one processor and a memory coupled to the processor, the first server being in communication with at least one second server, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising:

a program module that directs the first server to function in a specified manner to provide load-balancing functionality upon receiving a request for an asset from a client, the program module including instructions for;

determining prior to responding to the client if said first server has said asset and if said first server has sufficient resources to stream said asset;

if said first server does not have said asset, or said first server does not have sufficient resources to stream said asset, attempting, prior to streaming said asset, to identify a second server having said asset and sufficient resources to stream said asset;

informing the client to send the request to said identified second server; and

determining if said asset is a hot asset, and if said asset is determined to he a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

37. A computer program product for use in conjunction with a first server having at least one processor and a memory coupled to the processor, the first server being in communication with at least one second server, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising:

a program module that directs the first server to function in a specified manner to provide load-balancing functionality upon receiving a request for an asset from a client wherein said first server has said asset and sufficient resources to stream said asset, the program module including instructions for;

determining prior to responding to the client if said first server has a first server load level less than a load threshold value; and

if said first server has a load level greater than a load threshold level, said method further comprising;

attempting, prior to streaming said asset, to locate a second server having said asset, sufficient resources to stream said asset, and a second server load level less than said first server load; and

(i) informing the client to send the request to said second server if said second server is located;

or (ii) streaming said request if said server is not located; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server. and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

38. The computer program product of claim 37, wherein said program module further includes instructions for periodically:

calculating the first server load level; and

sending or broadcasting the first server load level to other servers including said second server.

39. A computer program product for use in conjunction with a first server having at least one processor and a memory coupled to the processor, the first server being in communication with at least one second server, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism, comprising:

a program module that directs the first server to function in a specified manner to provide load-balancing functionality upon receiving a request for an asset from a client, the program module including instructions for;

determining prior to responding to the client if said first server has said asset and if said first server has sufficient resources to stream said asset;

if said first server has said asset and sufficient resources to stream said asset, determining if said first server has a load level less than a load threshold value; and

if said first server has a load level greater than a load threshold level, attempting, prior to streaming said asset, to find a second server having said asset;

sufficient resources to stream said asset, and a second server load level less than said first server load level; and

(i) informing the client to send the request to the second server if said second server is located;

or (ii) streaming said request if said second server is not located; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associatcd with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) conying the asset to the second server and streaming the prefix of the asset from the second server.

40. The computer program product of claim 39, wherein said program module further includes instructions for:

periodically calculating said first server load level and sending or broadcasting said first server load level to other servers including the second server.

41. A server system for time-based media streaming comprising:

a plurality of servers coupled for communication with each other, including a first server, said first server comprising;

a first computer-readable storage medium encoded with server information comprising asset information that is replicated in server information stored in at least one other server; and

a computer program mechanism, the computer program mechanism, comprising;

a program module that directs the first server to function in a specified manner to provide load-balancing functionality upon receiving a request for an asset from a client and prior to responding to the client, the program module including instructions for;

determining if said first server has said asset;

determining if said first server has sufficient resources to stream said asset;

if said first server does not have said asset or said first server does not have sufficient resources to stream said asset, informing the client to send said request to a second server, the second server having said asset and sufficient resources to stream said asset;

or

if said first server has said asset and sufficient resources to stream said asset, determining if said first server has a load level less than a load threshold value;

if said first server has a first server load level less than a load threshold value, streaming said asset; and

if said first server has a load level greater than a load threshold level, attempting, prior to streaming said asset, to locate a second server having said asset, sufficient resources to stream said asset, and a second server load level less than said first server load level;

(i) forwarding said request, prior to streaming said asset, if said second server is located; and

(ii) streaming said request if said second server is not located; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

42. The server system for time-based media streaming in claim 41, wherein said time-based media streaming comprises maintaining streaming of said media within a predetermined play rate range.

43. The server system for time-based media streaming in claim 41, wherein said time-based media streaming comprises maintaining streaming of said media within a predetermined play rate range, and maintaining said media play within a predetermined delay relative to an absolute time reference.

44. The server system for time-based media streaming in claim 41, wherein said time-based media streaming comprises real-time media streaming.

45. A method for operating a time-base accurate asset streaming business, said method comprising:

operating a plurality of servers each configured to receive and service requests for assets from clients of said business, said operating step comprising;

operating a first server to determine if a received request for an asset from a client should be forwarded to another server prior to responding to the client based on existence of at least one of a plurality of conditions consisting of;

that the asset does not exist at the server;

that sufficient resources do not exist to stream the asset in real time from the server; and

that a current load on the server is over a threshold limit;

operating said first server to find a second server in the server system in response to the determination that the request should be forwarded;

operating said first server to inform the client to send the request to the second server after the second server is found; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

46. The method for operating a time-base accurate asset streaming business as in claim 45, wherein said operating a plurality of servers further comprises operating each server to periodically calculate at least one load factor indicating a current load at the server and send or broadcast the at least one load factor to other servers.

47. The method for operating a time-based accurate asset streaming business as in claim 45, wherein said determining if said first server has sufficient resources to stream said asset in real-time includes performing first server load assessment.

48. The method for operating a time-based accurate asset streaming business as in claim 45, wherein said operating a plurality of servers further comprises storing assets at each server and operating each server to broadcast an asset insert message to other servers in response to an asset being added at the server and to broadcast an asset delete message to other servers in response to an asset being removed at the server.

49. The method for operating a time-based accurate asset streaming business as in claim 48, wherein said operating a plurality of servers further comprises operating each server to communicate a list of assets stored at the server to other servers.

50. The method for operating a time-based accurate asset streaming business as in claim 45, wherein said operating a plurality of servers further comprises operating each server to determine if an asset is hot based on a number of received requests for the asset within a predetermined period and to copy the asset or a prefix of the asset to at least one other server in response to the determination that the asset is hot.

51. The method for operating a time-based accurate asset streaming business as in claim 45, wherein said operating a plurality of servers further comprises operating each server to request transfer of an asset from another server in response to receiving a request for the asset and a determination that a part of the asset is stored at the server.

52. In a server system comprising a plurality of servers for time-based media streaming, a method performed by a first server in the server system, the method comprising:

receiving a request for an asset from a client coupled to the plurality of servers;

determining if the request should be forwarded to another server before responding to the client based whether at least one of a plurality of conditions exists, said plurality of conditions consisting of;

that the asset does not exist at the server;

that sufficient resources do not exist to stream the asset from the server; and

that a current load on the server is over a threshold limit;

finding a second sewer in the server system in response to the determination that the request should be forwarded;

causing the request from the client to be sent to the second server after the second sewer is found; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server and (b) copying the asset to the second server and streaming the prefix of the asset from the second server.

53. A method according to claim 52, further comprising periodically calculating at least one load factor corresponding to the first sewer and sending or broadcasting the at least one load factor to other servers in the server system, and storing load factors received from other servers, wherein finding a second sewer further comprises finding the second sewer that is the least loaded among sewers having the asset according to load factors stored at the first server.

54. A method according to claim 53, wherein causing the request from the client to be sent to the second sewer comprises informing the client to send the request to the second server.

55. A method according to claim 53, wherein the first server in the server system stores a hot asset count value, the method further comprising determining if a number of received requests for the asset during a predetermined period exceeds the hot asset count value and replicating the asset or a variable length prefix of the asset to another server in response to a determination that the number exceeds the hot asset count value.

56. A method according to claim 52, wherein the first server comprises a computer storage medium storing therein asset information, the method further comprising sending or broadcasting an asset insert message to other sewers in response to an asset being inserted in the asset information and to broadcasting an asset delete message to other sewers in response to an asset being removed from the asset information.

57. A method for time-based media streaming, the method comprising:

at a first server in a cluster, receiving a request for a streaming session of a time-based media asset from a client, the asset having a prefix portion and a remainder portion;

at the first server, identifying a second server in the cluster having at least the prefix portion of the asset;

at the first server, receiving load information from the second server;

determining at the first server, based on load information associated with the first server and the received load information from the second server, whether to forward the request to the second server before responding to the client;

from the first server, transmitting a message to the client instructing the client to request streaming of the prefix of the time-based media asset from the second server; and

determining if said asset is a hot asset, and if said asset is determined to be a hot asset, then copying information associated with said asset to said second server including copying at least one of;

(a) a prefix of said asset to said second server, and (b) copying the asset to the second server and stream in the prefix of the asset from the second server.

58. The method of claim 57, wherein the message transmitted from the first server to the client comprises the name of the second server in the cluster.

59. The method of claim 58, wherein the second server is determined to be a server in the cluster with a least load with respect other servers in the cluster.

60. The method of claim 58, further comprising receiving at the second server a setup call from the client.

61. The method of claim 60, wherein the setup call comprises a message from the client causing the second server to allocate resources for a stream and start a time-based streaming session with the client.