SCALABLE FLOW TRANSPORT AND DELIVERY NETWORK AND ASSOCIATED METHODS AND SYSTEMS

• US 20100014528A1
• Filed: 07/21/2009
• Published: 01/21/2010
• Est. Priority Date: 07/21/2008
• Status: Active Grant

• Current Assignee: LTN Global Communications Inc.
• Original Assignee: Livetimenet, Inc.
• Inventors: Javadi, Yousef, Khan, Malik, Miskin-Amir, Michal, Amir, Yair, Stanton, Jonathan
• Granted Patent: US 8,619,775 B2
• Time in Patent Office: Days
• US Class Current: 370/400
• CPC Class Codes: H04L 45/04 Interdomain routing, e.g. hierarchical routing
• CPC Class Codes: H04L 45/125 based on throughput or bandwidth
• CPC Class Codes: H04L 45/306 Route determination based on the nature of the carried application
• CPC Class Codes: H04L 45/38 Flow based routing
• CPC Class Codes: H04L 45/64 using an overlay routing layer
• CPC Class Codes: H04L 47/2416 Real-time traffic
• CPC Class Codes: H04L 47/283 in response to processing delays, e.g. caused by jitter or round trip time [RTT]
• CPC Class Codes: H04L 47/805 QOS or priority aware

Abstract

A Flow Transport and Delivery Network (FTDN) supports delivery and transport of flows to destination nodes over parallel overlay networks. Overlay nodes executing overlay processes create and maintain one or more overlay networks. A plurality of sites are connected to each other via the parallel overlay networks with each site comprising one or more transit routers that route the flows to the destination nodes.

Claims

1. A network for scalable delivery of multiple flows amongst a plurality of nodes, comprising:

A plurality of overlay nodes, executing a plurality of overlay processes, said plurality of overlay processes managing data delivery via the multiple flows within parallel overlay networks; and

a communication system that shares state information amongst at least two overlay processes that maintain at least two overlay networks, said state information being associated with at least two identified flows.

2. The network of claim 1, wherein said multiple flows comprise multiple live flows.

3. The network of claim 1, wherein the plurality of overlay processes share a network resource based on an a-priori known information without communicating with each other.

4. The network of claim 1, wherein the plurality of overlay processes are assigned to a network resource based on gathered information about at least one of link status, resource utilization, or a parallel overlay networks'"'"' activity.

5. The network of claim 1, wherein the state information is shared between at least two overlay processes that maintain at least two overlay networks using a group communication service.

6. The network of claim 1, wherein the plurality of nodes comprise one or more destination nodes, wherein delivery of a flow to a destination node is in response to at least one service request received from the destination node.

7. The network of claim 1, wherein one or more of the plurality of overlay nodes comprise a dedicated routing device that operates under the control of one or more operating systems.

8. The network of claim 1, wherein the communication system provides access to said state information that is shared between at least two overlay processes that maintain at least two overlay networks.

9. The network of claim 1, wherein the communication system replicates said state information of one overlay process that maintains an overlay network by pushing the said state information to be shared by at least one other overlay process that maintains another overlay network.

10. The network of claim 1, wherein at least one of the overlay processes uses an overlay protocol that supports recovering an out-of-order packet associated with a flow and reordering said out-of-order packet.

11. The network of claim 10, wherein said out-of-order packet is recovered on a hop-by-hop basis.

12. The network of claim 1, wherein at least one of the overlay processes uses an overlay protocol that supports recovery of lost packets associated with the multiple flows over one hop upstream from another overlay node.

13. The network of claim 1, wherein at least one of the overlay processes uses an overlay protocol that does not support packet recovery.

14. The network of claim 1, wherein at least one of the overlay processes uses an overlay protocol that supports recovery of some lost packets.

15. The network of claim 1, wherein at least one of the overlay processes supports reliable message transport.

16. The network of claim 1 further including a service controller that controls at least one of a secure logging service, an audit service, or a flow replay service for a plurality of nodes.

17. The network of claim 1, wherein the communication system shares state information relating to allocated flows to a node amongst a plurality of overlay processes.

18. The network of claim 1 further comprising one or more access nodes.

19. The network of claim 18, wherein at least one of the overlay nodes comprises a transit router that executes the access node.

20. The network of claim 1, wherein the parallel overlay networks are assigned multiple IP addresses, each of which is claimed by an overlay node associated with the parallel overlay networks.

21. The network of claim 1, wherein the communication system shares state information relating to an allocated resource to an overlay node associated with the parallel overlay networks.

22. A network for scalable delivery of multiple flows amongst a plurality of nodes, comprising:

A plurality of overlay nodes run over an underlying IP network, executing a plurality of MOON processes, said plurality of MOON processes managing data delivery via the multiple flows within parallel overlay networks with three or more sites; and

a communication system that shares state information amongst at least two MOON processes that maintain at least two overlay networks, said state information being associated with at least two identified flows.

23. The network of claim 22, wherein said multiple flows comprise multiple live flows.

24. The network of claim 22, wherein the plurality of overlay processes share a network resource based on an a-priori known information without communicating with each other.

25. The network of claim 22, wherein the plurality of overlay processes are assigned to a network resource based on gathered information about at least one of link status, resource utilization, or a parallel overlay networks'"'"' activity.

26. The network of claim 22, wherein the state information is shared between at least two overlay processes that maintain at least two overlay networks using a group communication service.

27. The network of claim 22, wherein the plurality of nodes comprise one or more destination nodes, wherein delivery of a flow to a destination node is in response to at least one service request received from the destination node.

28. The network of claim 22, wherein one or more of the plurality of overlay nodes comprise a dedicated routing device that operates under the control of one or more operating systems.

29. The network of claim 22, wherein the communication system provides access to said state information that is shared between at least two overlay processes that maintain at least two overlay networks.

30. The network of claim 22, wherein the communication system replicates said state information of one overlay process that maintains an overlay network by pushing the said state information to be shared by at least one other overlay process that maintains another overlay network.

31. The network of claim 22, wherein at least one of the overlay processes uses an overlay protocol that supports recovering an out-of-order packet associated with a flow and reordering said out-of-order packet.

32. The network of claim 31, wherein said out-of-order packet is recovered on a hop-by-hop basis.

33. The network of claim 22, wherein at least one of the overlay processes uses an overlay protocol that supports recovery of lost packets associated with the multiple flows over one hop upstream from another overlay node.

34. The network of claim 22, wherein at least one of the overlay processes uses an overlay protocol that does not support packet recovery.

35. The network of claim 22, wherein at least one of the overlay processes uses an overlay protocol that supports recovery of some lost packets.

36. The network of claim 22, wherein at least one of the overlay processes supports reliable message transport.

37. The network of claim 22 further including a service controller that controls at least one of a secure logging service, an audit service, or a flow replay service for a plurality of nodes.

38. The network of claim 22, wherein the communication system shares state information relating to allocated flows to a node amongst a plurality of overlay processes.

39. The network of claim 22 further comprising one or more access nodes.

40. The network of claim 39, wherein at least one of the overlay nodes comprises a transit router that executes the access node.

41. The network of claim 22, wherein the parallel overlay networks are assigned multiple IP addresses, each of which is claimed by an overlay node associated with the parallel overlay networks.

42. The network of claim 22, wherein the communication system shares state information relating to an allocated resource to an overlay node associated with the parallel overlay networks.

43. A transit router that routes multiple flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more overlay processes associated with parallel overlay networks; and

a communication system that shares state information amongst at least two overlay processes that maintain at least two overlay networks, said state information being associated with at least two identified flows.

44. The transit router of claim 43, wherein the communication service provides access to state information that is shared between at least two overlay processes that maintain at least two overlay networks.

45. The transit router of claim 43 wherein a message associated with the flows is stored in at least one of system memory, disk drive, a persistent storage device associated with the transit router, or a host collocated with the transit router.

46. The transit router of claim 43, further including a coordinator process that makes at least one determination regarding coordination of the multiple flows locally in the transit router based on global state information.

47. The transit router of claim 46, wherein said at least one determination relates to at least one of admission of a flow to an overlay network, allocating an overlay network to an overlay process or allocating an overlay process to a node.

48. The transit router of claim 47, wherein admission of a flow to an overlay network is based on a priority information.

49. A transit router that routes multiple flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

a communication system that shares state information amongst at least two MOON processes that maintain at least two overlay networks, said state information being associated with at least two identified flows.

50. The transit router of claim 49, wherein the communication service provides access to state information that is shared between at least two overlay processes that maintain at least two overlay networks.

51. The transit router of claim 49 wherein a message associated with the flows is stored in at least one of transit router'"'"'s system memory or in a host collocated with the transit router.

52. The transit router of claim 49, further including a coordinator process that makes at least one determination regarding coordination of the multiple flows locally in the transit router based on global state information.

53. The transit router of claim 52, wherein said at least one determination relates to at least one of admission of a flow to an overlay network, allocating an overlay network to an overlay process or allocating an overlay process to a node.

54. The transit router of claim 53, wherein admission of a flow to an overlay network is based on a priority information.

55. A transit router that routes one or more flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more overlay processes associated with parallel overlay networks; and

a communication system that replicates state information of one overlay process that maintains an overlay network by pushing the state information to be shared by at least one other overlay process that maintains another overlay network.

56. The transit router of claim 55, wherein the parallel overlay networks support multiple flows.

57. The transit router of claim 56 further including a coordinator process that uses a consistent replicated data sharing system to make global determinations locally.

58. The transit router of claim 57, wherein the consistent replicated data sharing system comprises at least one of a Congruity or Replicated Hash Table.

59. The transit router of claim 57, wherein said determinations relate to at least one of admission of a flow to an overlay network, allocating an overlay network to an overlay process or allocating an overlay node to an overlay process.

60. A transit router that routes one or more flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

a communication system that replicates state information of one MOON process that maintains an overlay network by pushing the state information to be shared by at least one other MOON process that maintains another overlay network.

61. The transit router of claim 60, wherein the parallel overlay networks support multiple flows.

62. The transit router of claim 61 further including a coordinator process that uses a consistent replicated data sharing system to make global determinations locally.

63. The transit router of claim 62, wherein the consistent replicated data sharing system comprises at least one of a Congruity or Replicated Hash Table.

64. The transit router of claim 62, wherein said determinations relate to at least one of admission of a flow to an overlay network, allocating an overlay network to an overlay process or allocating an overlay node to an overlay process.

65. A transit router that routes one or more flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more overlay processes associated with parallel overlay networks; and

a group communication service that replicates state information amongst at least two overlay processes that maintain at least two overlay networks.

66. The transit router of claim 65, wherein the parallel overlay networks support multiple flows.

67. The transit router of claim 66, wherein the group communication service replicates state information amongst at least two overlay processes running in at least two transit routers associated with the parallel overlay networks.

68. The transit router of claim 66, wherein the group communication service replicates global state information amongst at least two overlay processes running in at least two super nodes associated with the parallel overlay networks.

69. The transit router of claim 66, wherein multiple overlay nodes share the same instance of the group communication service.

70. The transit router of claim 66, wherein a group communication service instance coordinates replication of state information in one or more transit routers within each super node, and another group communication service instance coordinates replication of state information between super nodes.

71. A transit router that routes one or more flows within a network having a plurality of nodes, comprising:

One or more processing units, said one or more processing units running one or more MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

a group communication service that replicates state information amongst at least two MOON processes that maintain at least two overlay networks.

72. The transit router of claim 71, wherein the parallel overlay networks support multiple flows.

73. The transit router of claim 72, wherein the group communication service replicates state information amongst at least two overlay processes running in at least two transit routers associated with the parallel overlay networks.

74. The transit router of claim 72, wherein the group communication service replicates global state information amongst at least two overlay processes running in at least two super nodes associated with the parallel overlay networks.

75. The transit router of claim 72, wherein multiple overlay nodes share the same instance of the group communication service.

76. The transit router of claim 72, wherein a group communication service instance coordinates replication of state information in one or more transit routers within each super node, and another group communication service instance coordinates replication of state information between super nodes.

77. A framework for delivery of one or more flows within a network having a plurality of nodes, comprising:

One or more processing units that run a first overlay process under the control of a first operating system to perform a function in a first overlay network, and a second overlay process under the control of a second operating system to perform a function in a second overlay network, wherein said first overlay network and said second overlay network comprise parallel overlay networks; and

a communication controller that provides access to state information that is shared between the first overlay process and the second overlay process.

78. The framework of claim 77, wherein the parallel overlay networks support multiple flows.

79. The framework of claim 78, wherein at least one of the first operating system and the second operating system is executed in a Virtual Machine.

80. The framework of claim 78, wherein the first operating system is different from the second operating system.

81. A framework for delivery of one or more flows within a network having a plurality of nodes, comprising:

One or more processing units that run a first MOON process under the control of a first operating system to perform a function in a first MOON, and a second MOON process under the control of a second operating system to perform a function in a second MOON, wherein said first MOON and said second MOON comprise parallel overlay networks with three or more sites on top of an underlying IP network; and

a communication controller that provides access to state information that is shared between the first overlay process and the second overlay process.

82. The framework of claim 81, wherein the parallel overlay networks support multiple flows.

83. The framework of claim 82, wherein at least one of the first operating system and the second operating system is executed in a Virtual Machine.

84. The framework of claim 82, wherein the first operating system is different from the second operating system.

85. A system for delivery of content within a network that has one or more destination nodes, comprising:

A plurality of sites, each having at least one transit router, said at least one transit router executing a plurality of overlay processes that are associated with parallel overlay networks; and

an inter-site network that delivers content to the destination nodes via plurality of live flows; and

a service controller that controls delivery of content to one or more destination nodes based on a service request received from destination nodes, said service requests relating to state information shared amongst at least two overlay processes that maintain at least two overlay networks.

86. The system of claim 85, wherein the at least one transit router recovers lost packets based on an attribute associated with a delivered content, wherein set attribute.

87. The system of claim 86, wherein the attribute relates to an I frame or a P frame of an MPEG content

88. The system of claim 85, wherein the service controller supports at least one of interactive services or flow replay.

89. The system of claim 85, wherein the underlying network comprises the Internet.

90. The system of claim 85, wherein the content comprises at least one of a video stream or a game stream.

91. A system for delivery of content within a network that has one or more destination nodes, comprising:

A plurality of sites, each having at least one transit router, said at least one transit router executing a plurality of MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

an inter-site network that delivers content to the destination nodes via plurality of live flows; and

a service controller that controls delivery of content to one or more destination nodes based on a service request received from destination nodes, said service requests relating to state information shared amongst at least two MOON processes that maintain at least two MOONs.

92. The system of claim 91, wherein the at least one transit router recovers lost packets based on an attribute associated with a delivered content, wherein set attribute.

93. The system of claim 92, wherein the attribute relates to an I frame or a P frame of an MPEG content

94. The system of claim 91, wherein the service controller supports at least one of interactive services or flow replay.

95. The system of claim 91, wherein the underlying network comprises the Internet.

96. The system of claim 91, wherein the content comprises at least one of a video stream or a game stream.

97. A system for delivery of one or more flows within a network that has one or more destination nodes, comprising:

A plurality of sites each having at least one transit router, said at least one transit router executing a plurality of overlay processes that are associated with parallel overlay networks; and

an inter-site network that delivers flows to the destination nodes;

said flows being delivered over a plurality of physical links that are associated with a plurality of bandwidth costs; and

one or more nodes executing one or more quality measurement processes associated with one or more flows, wherein the one or more flows are delivered to the destination nodes over links selected based on at least one of bandwidth cost and quality measurement.

98. The system of claim 97, wherein the parallel overlay networks support multiple live flows.

99. The system of claim 98, wherein the quality measure relates to at least one of latency, jitter or link reliability.

100. The system of claim 98, wherein the flows are delivered to the destination nodes based on at least one of a routing weight, available bandwidth or a hop-count.

101. The system of claim 98, wherein each overlay network is assigned addresses such as IP addresses, each of which is claimed by one overlay node.

102. The system of claim 98 further including a communication system that shares state information amongst at least two overlay processes that maintain at least two overlay networks.

103. The system of claim 102, wherein the state information relates to a resource allocated to an overlay node.

104. The system of claim 102, wherein the communication system shares information related to a flow allocated to an overlay networks.

105. The system of claim 102, wherein the communication system shares information related to a flow delivered to a destination node.

106. The system of claim 97, wherein at least one transit router selects links to optimize bandwidth costs while ensuring that the routing tree maintains a defined quality measure.

107. A system for delivery of one or more flows within a network that has a one or more destination nodes, comprising:

A plurality of sites each having at least one transit router, said at least one transit router executing a plurality of MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

an inter-site network that delivers flows to the destination nodes;

said flows being delivered over a plurality of physical links that are associated with a plurality of bandwidth costs; and

one or more nodes executing one or more quality measurement processes associated with one or more flows, wherein the one or more flows are delivered to the destination nodes over links selected based on at least one of bandwidth cost and quality measurement.

108. The system of claim 107, wherein the parallel overlay networks support multiple live flows.

109. The system of claim 108, wherein the quality measure relates to at least one of latency, jitter or link reliability.

110. The system of claim 108, wherein the flows are delivered to the destination nodes based on at least one of a routing weight, available bandwidth or a hop-count.

111. The system of claim 108, wherein each overlay network is assigned IP addresses, each of which is claimed by one overlay node.

112. The system of claim 108 further including a communication system that shares state information amongst at least two overlay processes that maintain at least two overlay networks.

113. The system of claim 112, wherein the state information relates to a resource allocated to an overlay node.

114. The system of claim 112, wherein the communication system shares information related to a flow allocated to an overlay networks.

115. The system of claim 112, wherein the communication system shares information related to a flow delivered to a destination node.

116. The system of claim 107, wherein at least one transit router selects links to optimize bandwidth costs while ensuring that the routing tree maintains a defined quality measure.

117. A system for delivery of one or more flows within a network having a plurality of nodes, comprising:

At least two transit routers, each executing one or more overlay processes that are associated with parallel overlay networks; and

a pairing of each shadow overlay process to a corresponding existing overlay process such that they execute on different transit routers; and

a fault-tolerant service that monitors the health of overlay processes and notifies the shadow overlay process of the failure of the corresponding overlay process;

118. The system of claim 117, further comprising a packet duplicator service that delivers a duplicate copy of packets that arrive at an overlay process to the corresponding shadow overlay process and wherein the shadow overlay process receives and temporarily stores the packets that arrive at it.

119. The system of claim 118, wherein the parallel overlay networks support multiple live flows.

120. A system for delivery of one or more flows within a network having a plurality of nodes, comprising:

At least two transit routers, each executing one or more MOON processes that are associated with parallel overlay networks with three or more sites and run over an underlying IP network; and

a pairing of each shadow MOON process to a corresponding existing MOON process such that they execute on different transit routers; and

a fault-tolerant service that monitors the health of MOON processes and notifies the shadow MOON process of the failure of the corresponding MOON process;

121. The system of claim 120, further comprising a packet duplicator service that delivers a duplicate copy of packets that arrive at an overlay process to the corresponding shadow overlay process and wherein the shadow overlay process receives and temporarily stores in system memory the packets that arrive at it.

122. The system of claim 121, wherein the parallel overlay networks support multiple live flows.