Systems and methods for communicating data over parallel data paths

• US 9,621,384 B2
• Filed: 02/22/2005
• Issued: 04/11/2017
• Est. Priority Date: 02/19/2004
• Status: Active Grant

• Current Assignee: Georgia Tech Research Corporation (University System of Georgia)
• Original Assignee: Georgia Tech Research Corporation (University System of Georgia)
• Inventors: Sivakumar, Raghupathy, Hsieh, Hung-Yun
• Primary Examiner(s): Hsu, Alpus H
• Assistant Examiner(s): Houshmand, Hooman
• Application Number:
  US11/063,284
• Publication Number:
  US 20050185621A1
• Time in Patent Office: 4,431 Days
• Field of Search: None
• CPC Class Codes: H04L 1/1607 Details of the supervisory signal
• CPC Class Codes: H04L 1/1887 Scheduling and prioritising arrangements
• CPC Class Codes: H04L 25/14 Channel dividing arrangements , i.e. in which a single bit stream is divided between several baseband channels and reassembled at the receiver
• CPC Class Codes: H04L 45/24 Multipath
• CPC Class Codes: H04L 47/27 Evaluation or update of window size, e.g. using information derived from acknowledged [ACK] packets

Abstract

Systems and methods for the communication of data over a plurality of parallel communication paths are provided. Embodiments of the parallel communications systems and methods may discover, characterize, and leverage multiplicity of resources in various network elements to provide network applications with a desired communication objective and level of performance. The systems and methods may dynamically adapt to changes in the network resources to continuously provide the desired communication performance.

Claims

1. A method for communicating data over parallel data paths comprising:

characterizing each of a plurality of parallel data paths by repeatedly determining path properties, the plurality of parallel data paths defining a single virtual connection between a sending node and receiving node, wherein at least one of the plurality of parallel data paths comprises a plurality of intermediate nodes; and

scheduling the transmission of data across the plurality of parallel data paths to meet an objective function, the scheduling based on the path properties.

2. The method of claim 1, wherein the step of scheduling the transmission data further includes:

scheduling a first transmission of data along a first path of the plurality of parallel data communication paths to be received at the receiving node; and

scheduling a second transmission of data along a second path of the plurality of parallel data communication paths, the second transmission of data scheduled to be received at the receiving node in a predetermined sequence in relation to a first packet.

3. The method of claim 1, further including:

reassembling the received first and second transmissions of data at the receiving node only upon;

receiving the first and second transmissions out of the predetermined sequence; and

determining that a higher layer protocol requires the first and second transmissions be delivered in the predetermined sequence.

4. The method of claim 1, further including:

predicting an arrival time of data to be transmitted over each of the plurality of parallel data communication paths.

5. The method of claim 4, further including:

predicting the arrival probability of data to be transmitted over each of the plurality of parallel data communication paths.

6. The method of claim 4, further including:

synchronizing the arrival time of data at the receiving node within a predetermined degree of synchronization.

7. The method of claim 1, further including:

broadcasting a data packet comprising a request for a receiving parallel communication transceiver to transmit an acknowledgment packet; and

receiving an acknowledgement packet from the receiving parallel communication transceiver.

8. The method of claim 7, further including:

abstracting at least one path to the receiving parallel communication transceiver.

9. The method of claim 1, further including:

monitoring data traffic to discover a sending parallel communication transceiver and a receiving parallel communication transceiver; and

abstracting at least one path between the sending parallel communication transceiver and the receiving parallel communication transceiver.

10. The method of claim 1, further including:

decoupling network functionalities into per-path functionalities and per-connection functionalities.

11. The method of claim 10, further including:

creating a session based on a trigger, the session including a core engine for maintaining the per-path functionalities and at least one path engine for maintaining the per-path functionalities.

12. The method of claim 10, further including:

maintaining a data structure for each of the plurality of parallel data paths.

13. The method of claim 10, wherein the per-path functionalities are selected from the group consisting of:

the estimation of an available rate, an estimation of delay, an estimation of loss, and congestion control.

14. The method of claim 10, wherein the per-connection functionalities are selected from the group consisting of:

flow control, connection management, and connection reliability.

15. The method of claim 1, wherein the objective function is providing the virtual connection with one or more levels of:

throughput, bandwidth, delay, jitter, loss rate, security, reliability, resilience, or cost.

16. The method of claim 1, wherein the path properties are a measurement of one or more of the instantaneous, mean, and distribution of the throughput, bandwidth, delay, jitter, loss rate, cost, security, reliability, resilience, or cost for data transmitted over a particular path.

17. The method of claim 1, wherein the step of scheduling is further based on an awareness of a higher layer network protocol or an application.

18. The method of claim 1, further including:

repeatedly transmitting packets to a plurality of known parallel communication transceivers for determining an availability of the known parallel communication transceivers to receive the transmission of the data.

19. The method of claim 1, further including:

instructing an intermediate parallel communication transceiver to direct data packets to other known parallel communication transceivers using a routing standard selected from the group consisting of;

IP-in-IP encapsulation, minimal IP encapsulation, and source routing.

20. The method of claim 1, wherein the step of scheduling the transmission of data across the plurality of parallel data paths further includes:

scheduling the data for transmission to the receiving node, the receiving node selected from the group consisting of;

a parallel communication transceiver and a destination endpoint.

21. The method of claim 1, wherein characterizing each of a plurality of parallel data paths further comprises adding a header to a packet, wherein the header comprises information used to deductively determine the path properties.

22. The method of claim 1, wherein in scheduling the transmission of data across the plurality of parallel data paths is based on the path properties to minimize out of order receipt.

23. A system for communicating data over parallel data paths comprising:

a processor configured to fetch and execute instructions from an instruction execution system, the instructions comprising;

instructions for characterizing each of a plurality of parallel data paths by repeatedly determining path properties, the plurality of parallel data paths defining a single virtual connection between a sending node and receiving node, wherein at least one of the plurality of parallel data paths comprises a plurality of intermediate nodes; and

a memory, coupled to the processor, having instructions, embodied thereon for scheduling the transmission of data across the plurality of parallel data paths to meet an objective function, the scheduling based on the path properties, the instructions for scheduling the transmission of data further including;

instructions, embodied on the memory, for scheduling a first transmission of data along a first path of the plurality of parallel data communication paths to be received at the receiving node; and

instructions, embodied on the memory, for scheduling a second transmission of data along a second path of the plurality of parallel data communication paths, the second transmission of data scheduled to be received at the receiving node in a predetermined sequence in relation to a first packet.

24. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for reassembling the received first and second transmissions of data at the receiving node only upon;

receiving the first and second transmissions out of the predetermined sequence; and

determining that a higher layer protocol requires the first and second transmissions be delivered in the predetermined sequence.

25. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for predicting an arrival time of data to be transmitted over each of the plurality of parallel data communication paths.

26. The system of claim 25, the instructions further including:

instructions, embodied on the memory, for predicting the arrival probability of data to be transmitted over each of the plurality of parallel data communication paths.

27. The system of claim 25, the instructions further including:

instructions, embodied on the memory, for synchronizing the arrival time of data at the receiving node within a predetermined degree of synchronization.

28. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for broadcasting a data packet comprising a request for a receiving parallel communication transceiver to transmit an acknowledgment packet; and

instructions, embodied on the memory, for receiving an acknowledgement packet from the receiving parallel communication transceiver.

29. The system of claim 28, the instructions further including:

instructions, embodied on the memory, for abstracting at least one path to the receiving parallel communication transceiver.

30. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for monitoring data traffic to discover a sending parallel communication transceiver and a receiving parallel communication transceiver; and

instructions, embodied on the memory, for abstracting at least one path between the sending parallel communication transceiver and the receiving parallel communication transceiver.

31. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for decoupling network functionalities into per-path functionalities and per-connection functionalities.

32. The system of claim 31, the instructions further including:

instructions, embodied on the memory, for creating a session based on a trigger, the session including a core engine for maintaining the per-path functionalities and at least one path engine for maintaining the per-path functionalities.

33. The system of claim 31, the instructions further including:

instructions, embodied on the memory, for maintaining a data structure for each of the plurality of parallel data paths.

34. The system of claim 23, wherein the instructions for scheduling the transmission of data across the plurality of parallel data paths are further based on an awareness of a higher layer network protocol or an application.

35. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for repeatedly transmitting packets to a plurality of known parallel communication transceivers for determining an availability of the known parallel communication transceivers to receive the transmission of the data.

36. The system of claim 23, the instructions further including:

instructions, embodied on the memory, for directing data packets to other known parallel communication transceivers using a routing standard selected from the group consisting of;

IP-in-IP encapsulation, minimal IP encapsulation, and source routing.

37. The system of claim 23, wherein the instructions for scheduling the transmission of data across the plurality of parallel data paths further includes:

instructions, embodied on the memory, for scheduling the data for transmission to the receiving node, the receiving node selected from the group consisting of;

a parallel communication transceiver and a destination endpoint.

38. A method for communicating data over parallel data paths comprising:

characterizing each of a plurality of parallel data paths by repeatedly determining path properties, the plurality of parallel data paths defining a single virtual connection between a sending node and receiving node, wherein at least one of the plurality of parallel data paths comprises a plurality of intermediate nodes; and

scheduling the transmission of data across the plurality of parallel data paths to meet an objective function, the scheduling based on the path properties, wherein the objective function is providing the virtual connection with one or more levels of bandwidth, delay or loss rate.