Method and apparatus for improving the performance of TCP and other network protocols in a communications network using proxy servers

• US 9,537,611 B2
• Filed: 11/06/2013
• Issued: 01/03/2017
• Est. Priority Date: 11/08/2012
• Status: Active Grant

• Current Assignee: Akamai Technologies, Inc.
• Original Assignee: Instart Logic Incorporated (Akamai Technologies, Inc.)
• Inventors: Zhovnirnovsky, Igor, Roy, Subhash
• Primary Examiner(s): Chery, Dady
• Application Number:
  US14/400,325
• Publication Number:
  US 20150236819A1
• Time in Patent Office: 1,154 Days
• Field of Search: 370/329, 370/216, 370/341, 370/346, 370/348
• US Class Current: 1/1
• CPC Class Codes: H04L 1/0041 Arrangements at the transmitter end
• CPC Class Codes: H04L 1/0057 Block codes H04L1/0061, H04L1/0064 take precedence
• CPC Class Codes: H04L 1/08 by repeating transmission, e.g. Verdan system H04L1/1858 and H04L1/189 take precedence
• CPC Class Codes: H04L 1/1819 with retransmission of additional or different redundancy
• CPC Class Codes: H04L 1/1845 Combining techniques, e.g. code combining
• CPC Class Codes: H04L 1/20 using signal quality detector
• 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: H04W 80/06 Transport layer protocols, e.g. TCP [Transport Control Protocol] over wireless transmission control protocol/Internet protocol [TCP/IP] or user datagram protocol [UDP] H04L69/16

Abstract

A method and apparatus that improves the performance of a data network by segmenting the TCP path and implementing a proprietary protocol (DPR™) over a network. Bandwidth is reduced and reliability improved by using an erasure coded algorithm to generate a predicted number of redundant coded packets used to reconstruct lost data packets. Coded packets are generated at the transmission side and the coded packets together with the raw data packets successfully sent over the channel are used to reconstruct lost raw data packets. The DPR™ erasure coding to adjust for packet loss in real time protocol provides a multiplexed tunnel for a multiplicity of TCP sessions from a client to a cloud proxy. DPR™ implements congestion management, flow control, reliability, and link monitoring. Other network protocols (such as UDP) are supported with a reliability protocol based upon network coding that improves the transmission reliability.

Claims

1. A method for transmitting a plurality of blocks of data packets in a digital data sequence, comprising:

receiving at a first transceiver a first block of data packets for transmission through a data communications channel to a second transceiver;

duplicating the first block of data packets and storing the packets to make a stored copy of the first block of data packets;

transmitting the first block of data packets from the first transceiver through the data communications channel to the second transceiver;

determining a first set of coded data packets from the stored copy of the first block of data packets;

transmitting the first set of coded data packets from the first transceiver through the data communications channel to the second transceiver;

receiving an indication of whether the coded data packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct one or more lost data packets;

in the event the coded data packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, reconstructing the one or more lost data packets; and

in the event the coded packets received by the second transceiver are not sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, transmitting a second set of coded data packets.

2. The method as in claim 1, wherein an additional set of coded data packets is transmitted until the coded packets received by the second-transceiver are sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets.

3. The method as in claim 1, wherein the data communications channel is the Internet.

4. The method as in claim 1, wherein a number of coded data packets in the first set of coded data packets is based at least in part on an estimated number of data packets lost during transmission.

5. The method as in claim 1, wherein coding the first set of coded data packets is performed using erasure coding.

6. The method as in claim 1, wherein coding the first set of coded data packets is performed using forward error correction.

7. The method as in claim 1, wherein in the event packet loss is greater than expected, a number of coded data packets transmitted in a first transmission of coded data packets for a second block of data packets is greater than a number of coded data packets in the first set of coded data packets for the first block of data packets.

8. The method as in claim 1, wherein in the event packet loss is equal to that expected, a number of coded data packets transmitted in a first transmission of coded data packets for a second block of data packets equals a number of coded data packets in the first set of coded data packets for the first block of data packets.

9. The method as in claim 1, further comprising:

receiving with the second transceiver all or some of the first block of data packets and the first set of coded data packets from the first transceiver;

counting received packets of the first block of data packets and the first set of coded data packets to generate packet loss information; and

transmitting the packet loss information from the second transceiver through the data communications channel to the first transceiver.

10. The method as in claim 1, wherein the data communications channel is a full duplex communications channel.

11. The method of claim 1, wherein a number of coded data packets in the first set of coded data packets is inversely proportional to a quality or a quality score of the data communications channel.

12. The method as in claim 11, wherein the quality score of the data communications channel is assigned based on one or more of the following:

a data packet loss rate, an inter data packet arrival time, or a data communicational channel round trip time.

13. The method as in claim 1, wherein a number of coded data packets in the second set of coded data packets is equal to a number of lost data packets.

14. The method as in claim 1 wherein the method comprises use of a proxy server.

15. The method as in claim 1 wherein the coded data packets are generated using one or more of the following:

Reed-Solomon (RS) coding, Low Density Parity Check (LDPC) coding, Luby transform (LT) coding, Raptor coding, Network coding, or Online codes.

16. The method as in claim 1, wherein the second set of coded data packets comprises different coded data packets than the first set of coded data packets.

17. The method as in claim 1, wherein a data packet of the block of data packets or a data packet of the first set of coded data packets comprises an indication of whether it is coded.

18. The method as in claim 1, wherein in the event that a number of data packets received by the transceiver, coded or non-coded, is equal to a number of data packets in the block of data packets, the transceiver has received sufficient information to reconstruct lost data packets of the block of data packets.

19. A system for transmitting a plurality of blocks of data packets in a digital data sequence, comprising:

a first transceiver configured to;

receive a first block of data packets for transmission through a data communications channel to a second transceiver; and

transmit the first block of data packets through the data communications channel to the second transceiver; and

a processor configured to;

duplicate the first block of data packets and storing the packets to make a stored copy of the first block of data packets;

determine a first set of coded data packets from the stored copy of the first block of data packets, wherein the first transceiver is configured to transmit the first set of coded data packets through the data communications channel to the second transceiver;

receive an indication of whether the coded data packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct one or more lost data packets;

in the event the coded packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, reconstruct the one or more lost data packets; and

in the event the coded data packets received by the second-transceiver are not sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, transmit a second set of coded data packets.

20. A computer program product for transmitting a plurality of blocks of data packets in a digital data sequence, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving at a first transceiver a first block of data packets for transmission through a data communications channel to a second transceiver;

duplicating the first block of data packets and storing the packets to make a stored copy of the first block of data packets;

transmitting the first block of data packets from the first transceiver through the data communications channel to the second transceiver;

determining a first set of coded data packets from the stored copy of the first block of data packets;

transmitting the first set of coded data packets from the first transceiver through the data communications channel to the second transceiver;

receiving an indication of whether the coded data packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct one or more lost data packets;

in the event the coded packets received by the second transceiver are sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, reconstructing the one or more lost data packets; and

in the event the coded data packets received by the second-transceiver are not sufficient together with received data packets of the first block of data packets to reconstruct the one or more lost data packets, transmitting a second set of coded data packets.