Network transport accelerator

• US 20020107971A1
• Filed: 03/01/2001
• Published: 08/08/2002
• Est. Priority Date: 11/07/2000
• Status: Abandoned Application

• Current Assignee: Surgient Networks, Inc. (Quest Software, Inc.)
• Original Assignee: Surgient Networks, Inc. (Quest Software, Inc.)
• Inventors: Richter, Roger K., Wang, Ho, Bailey, Brian W.
• Application Number:
  US09/797,412
• Publication Number:
  US 20020107971A1
• Time in Patent Office: Days
• US Class Current: 709/231
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 41/5022 by giving priorities, e.g. assigning classes of service
• CPC Class Codes: H04L 41/509 wherein the managed service relates to media content delivery, e.g. audio, video or TV
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0882 Utilisation of link capacity
• 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 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 69/10 Streamlined, light-weight or high-speed protocols, e.g. express transfer protocol [XTP] or byte stream
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/161 Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
• CPC Class Codes: H04L 69/165 Combined use of TCP and UDP protocols; selection criteria therefor
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A network endpoint system receives requests delivered in packet format via a network. The system uses a transport accelerator at its front end, which performs all or some of the network protocol processing. The transport accelerator is directly connected to one or more processing units, which respond to the requests. The protocol processing may be partitioned between the transport accelerator and the processing units in a manner that best uses their different processing capabilities.

Claims

1. A network endpoint system for responding to requests delivered in packet form having a networking protocol via a network, comprising:

a transport accelerator unit having at least a network processor programmed to receive packets and to perform at least some processing of the network/transport protocol;

at least one processing unit programmed to receive the packets from the network processor and to respond to the requests; and

an interconnection medium for directly connecting the network processor to the processing unit.

2. The system of claim 1, wherein the interconnection medium is a bus.

3. The system of claim 1, wherein the interconnection medium is a switch fabric.

4. The system of claim 1, wherein the network is the Internet.

5. The system of claim 1, wherein the network is a private network.

6. The system of claim 1, wherein the transport accelerator performs only some tasks of network/transport protocol processing, and the processing unit performs the remaining tasks.

7. The system of claim 6, wherein the processing unit performs all tasks requiring state information.

8. The system of claim 1, wherein the transport accelerator is programmed to perform all protocol processing such that it passes data to the processing unit at the transport interface level.

9. The system of claim 1, wherein the network/transport protocol is the TCP/IP protocol.

10. The system of claim 1, wherein the network/transport protocol is the UDP/IP protocol.

11. The system of claim 1, wherein the network/transport protocol is at or below the RTP protocol.

12. The system of claim 1, wherein the transport accelerator also has a transport processor for sharing transport processing tasks with the network processor.

13. The system of claim 1, wherein the transport accelerator and the processing unit are physically separate devices.

14. The system of claim 1, wherein the system is implemented as a single chassis system.

15. The system of claim 1, wherein the endpoint system is a server system.

16. The system of claim 1, wherein the endpoint system is a client system.

18. The method of claim 17, wherein the network is the Internet.

19. The method of claim 17, wherein the network is a private network.

20. The method of claim 17, further comprising the step of dividing tasks of the network/transport protocol, such that the transport accelerator performs only some tasks of network/transport layer processing, and the processing unit performs the remaining tasks.

21. The method of claim 20, wherein the processing unit performs all tasks requiring state information.

22. The method of claim 17, wherein the transport accelerator is programmed to perform all protocol processing such that it passes data to the processing unit at the transport interface level.

23. The method of claim 17, wherein the network/transport protocol is the TCP/IP protocol.

24. The method of claim 17, wherein the network/transport protocol is the UDP/IP protocol.

25. The method of claim 17, wherein the network/transport protocol is the RTP protocol and all lower protocols.

26. The method of claim 17, wherein the transport accelerator performs checksum tasks.

27. The method of claim 17, wherein the transport accelerator performs header generation and verification tasks.

29. The device of claim 28, wherein the interconnection medium is a bus.

30. The device of claim 28, wherein the interconnection medium is a switch fabric.

31. The device of claim 28, wherein the interconnection medium is shared memory.

32. The device of claim 28, wherein the transport accelerator, the front end interface, and the back end interface are fabricated as a single circuit component.

33. The device of claim 28, wherein the transport accelerator performs only some tasks of network/transport protocol processing, namely, tasks not requiring state information.

34. The device of claim 28, wherein the transport accelerator is programmed to perform all protocol processing such that it delivers data from the back end interface at the transport interface level.

35. The device of claim 28, wherein the network/transport protocol is the TCP/IP protocol.

36. The device of claim 28, wherein the network/transport protocol is the UDP/IP protocol.

37. The device of claim 28, wherein the network/transport protocol is at or below the RTP protocol.

38. The device of claim 28, wherein the transport accelerator also has a transport processor for sharing transport processing tasks with the network processor.

39. The device of claim 28, wherein the transport processor and network processor are connected with an internal interconnection medium.

40. The device of claim 28, wherein the transport acceleration further has a bridge as the back end interface.

17. A method of processing network packets at a network endpoint system that responds to requests delivered in packet form having a networking protocol via a network, comprising the steps of:

directly connecting a transport accelerator, which has at least a network processor, to one or more processing units;

receiving the packets at the transport accelerator;

using the transport accelerator to perform at least some processing of the network/transport protocol;

delivering the packets to at least one processing unit; and

using the processing unit to respond to the requests.

28. A transport accelerator device for use at a network endpoint, comprising:

a network processor programmed to receive packets and to perform at least some processing of the network/transport protocol

a front end interface for connecting the transport accelerator to a network; and

a back end interface for connecting the transport accelerator to an interconnection medium.

41. A network connectable computing system, the system being configured to be connected on at least one end to a network, the system comprising:

at least one network connection configured to be coupled to the network;

a first system processor for performing system functionality;

a second system processor located in a data path between the network connection and the at first system processor; and

an interconnection between the at least one processor and the second system processor,

wherein the second system processor processes a portion of data packets provided to the system from the network and then forwards the data packets data packets to the remainder of the system so that the system functionality may be performed upon the data packets

42. The system of claim 41, wherein the second processor comprises a network processor.

43. The system of claim 42, wherein the network processor performs at least some protocol processing of the data packets.

44. The system of claim 42, further comprising a third system processor, the protocol processing of data packets being split between the network processor and the third system processor

45. The system of claim 44, wherein the first system processor, the network processor, and the third system processor communicate in a peer to peer environment across a distributed interconnect.

46. The system of claim 45, wherein the first system processor comprises an application processor, the system further comprising a storage processor.

47. The system of claim 41, wherein the network connectable computing system is a network endpoint system and the at least first system processor comprises an application processor, the system further comprising a storage processor.

48. The system of claim 47, wherein the interconnection is a switch fabric.

50. The method of claim 49, wherein the removing step offloads processing steps from the first system processor.

51. The method of claim 49, wherein the wherein the first system processor is a transport processor which performs additional protocol processing.

52. The method of claim 51, wherein after processing by the transport processor the data is forwarded to a second system processor.

53. The method of claim 49, wherein the first system processor is an application processor or a storage processor.

54. The method of claim 49, wherein the contextually meaningful data is an identifier.

55. The method of claim 49, further comprising providing at least one data packet having full header information to the first system processor and subsequently providing to the first system processor a plurality of data packets having the at least a portion of the data packet headers removed and replaced.

56. The method of claim 55, wherein the network connected computing system is a network endpoint system.

57. The method of claim 56, wherein the removing step accelerates the delivery of content from the network endpoint system.

59. The method of claim 58, wherein the first system engine is a transport engine, the transport engine performing additional protocol processing.

60. The method of claim 58, wherein the network interface engine performs all protocol processing.

61. The method of claim 58, wherein at least one initial data packet for a connection to the network endpoint system does not have lowered order layers replaced prior to being forwarded to the first system engine.

62. The method of claim 61, further comprising processing the lowered ordered layers within the first system engine to obtain a processor result, the additional data being used to identifier the processor result for use with subsequent data packets received after the at least one initial data packet.

63. The method of claim 61, wherein the first system engine is a transport engine, the transport engine performing additional protocol processing.

64. The method of claim 61, wherein the network interface engine performs all protocol processing.

65. The method of claim 61, wherein the network connected computing system is a content delivery system, the accelerated operation providing accelerated content delivery.

49. A method of operating a network connected computing system, comprising:

receiving data from a network;

analyzing the data with a network interface engine to decode incoming data packet headers;

removing at least a portion of the data packet headers of at least some data packets and replacing the removed headers with contextually meaningful data based upon the analysis of the data packet header; and

forwarding the data packet to at least a first system processor through a system interconnection after replacing the removed headers.

58. A method accelerating the operation of a network connected computing system, comprising:

receiving, in a network interface engine, data packets from a network, the data packets provided in a layered protocol;

analyzing a plurality of lower ordered layers of the data packets with the network interface engine;

replacing the lowered order layers of the data packets with additional data;

transmitting the data packet containing the additional data to at least a first system engine, the first system engine having accelerated operation due to processing the additional data as compared to processing the plurality of lower ordered layers.

66. A network endpoint system for performing endpoint functionality, the endpoint system comprising:

at least one system processor, the system processor performing endpoint processing functionality;

a distributed interconnect coupled to the at least one system processor; and

a network interface engine coupled to the distributed interconnect,

wherein the system is configured such that a data packet from a network may be processed by the network interface engine prior to being processed by the at least one system processor, the processing by the network interface engine comprising replacing at least a portion of lower ordered protocol layers with an identifier associated with the content of the removed lower ordered layers.

67. The network endpoint system of claim 66, the network endpoint system configured in a asymmetric staged pipelined processing systems.

68. The network endpoint system of claim 66, wherein the at least one system processor comprises at least one storage processor and at least one application processor.

69. The network endpoint system of claim 68, wherein the network interface engine comprises at least one network processor.

70. The network endpoint system of claim 69, wherein the network processor, the storage processor and the application processor operate in a peer to peer environment across the distributed interconnect.

71. The network endpoint system of claim 70, wherein the distributed interconnect is a switch fabric.

72. The network endpoint system of claim 66, wherein the network endpoint system is a content delivery system.

73. The network endpoint system of claim 72 wherein:

the network interface engine comprises at least one network processor;

the at least one system processor comprises at least one storage processor and at least one application processor, the storage processor being configured to interface with a storage system; and

the network processor, the storage processor and the application processor operate in a peer to peer environment across the distributed interconnect.

74. The network endpoint system of claim 73 wherein the distributed interconnect is a switch fabric.

75. The network endpoint system of claim 74, wherein the system is configured in a single chassis.

76. A method of operating a network endpoint system, comprising:

providing a network processor within the network endpoint system, the network processor being at an interface which couples the network endpoint system to a network;

processing data packets passing through the interface with the network processor;

removing portions of the data packets layers as part of the processing of the network processor; and

forwarding incoming network data from the network processor to a system processor which performs at least some endpoint functionality upon the data.

77. The method of claim 76 wherein incoming network data is forwarded to the system processor through a transport processor that performs at least some protocol processing.

78. The method of claim 76 wherein the network processor forwards at least some data packets without removing the portions of the data packets removed from other data packets.

79. The method of claim 78 wherein the network processor replaces the removed portions of the data packet layers with identifiers that identify the contents of the removed data packet layers.

80. The method of claim 78, wherein the at least some data packets in which the portions are not removed are one or more data packets that initialize a connection to the network endpoint system.

81. The method of claim 80 wherein the system is configured in a staged pipelined manner, a plurality of the stages of the system replacing layers of the data packets with identifiers.

82. The method of claim 78 wherein, further comprising performing split protocol processing in which the network processor performs only a portion of the protocol processing.

83. The method of claim 78 wherein the network endpoint system is a content delivery system.

84. The method of claim 78, wherein the content delivery system is configured in a peer to peer environment.

85. The method of claim 84 wherein peer to peer communications are provided across a switch fabric.

86. A network connectable computing system, comprising:

a first connection to receive data packets from a network;

a network interface engine comprising at least one network processor, the network processor coupled to the interface connection; and

a second connection to transmit data processed by the network interface engine,

wherein the at least one network processor analyzes the data packets and removes at least a portion of the headers of the data packets and replaces the removed portions with identifiers which may be utilized to reduced subsequent processor workloads.

87. The system of claim 86, wherein the network processor processes at least some data packets of a network connection without removing the headers.

88. The system of claim 86, wherein the system is an intermediate network node system.

89. The system of claim 88, wherein the system is a network switch.

90. The system of claim 86, wherein the system is a network endpoint system.

91. The system of claim 86, wherein the system is a network endpoint system having at least one server or at least one server card coupled to the second connection.

92. The system of claim 86, wherein the system is incorporated into a network interface card.

93. The system of claim 91, wherein the second connection is a distributed interconnection.

94. The system of claim 93, wherein the distributed interconnection is a switch fabric.

95. The system of claim 86, wherein the second connection is coupled to an asymmetric multi-processing system.

96. The system of claim 95, wherein the second connection is a distributed interconnection and the asymmetric multi-processing system includes a plurality of task specific processors.

97. The system of claim 96, wherein the distributed interconnection is a switch fabric and the task specific processors include storage or application processors.

98. The system of claim 97, wherein the task specific processors include storage and application processors.