System and method of reducing latency using adaptive retransmission timeouts

• US 8,468,196 B1
• Filed: 05/20/2010
• Issued: 06/18/2013
• Est. Priority Date: 05/20/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Roskind, James, Belshe, Michael A.
• Primary Examiner(s): MIRZA, ADNAN M
• Application Number:
  US12/783,797
• Time in Patent Office: 1,125 Days
• Field of Search: 709/203, 709/204
• US Class Current: 709/203
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 67/62 Establishing a time schedule for servicing the requests
• CPC Class Codes: H04L 69/28 Timers or timing mechanisms used in protocols

Abstract

In one aspect, a system and method are provided whereby latency in network communication protocols such as the TCP/IP suite of protocols is reduced by transmitting a new and second connection request from a sending device to a receiving device over a network based upon adaptively determined dynamic initial timeout values, where the dynamic initial timeout values are adaptively determined based upon data associated with one or more historical requests transmitted over the network by the sending device.

Claims

1. A method of transmitting packets over a network comprising:

adaptively determining, in an application running on a application layer in a first device, a dynamic initial timeout value with a processor, where the dynamic initial timeout value is adaptively determined based on a time within which one or more previous responses were received at the first device over a network in response to one or more historical requests, wherein the historical request is a non-pending and closed request;

transmitting a first request for first information from the first device to a second device over the network;

before a response is received from the second device to the first request, performing a comparison between the time that has elapsed since the first request was transmitted and the adaptively determined dynamic initial timeout value; and

,

transmitting, depending on the comparison and via the network, a new request for the first information to the second device, where the new request is distinct from and unrelated to the first request.

2. The method of claim 1, wherein the first request and the new request are network connection requests transmitted in accordance with the TCP/IP protocol.

3. The method of claim 1, wherein the first request and the new request are domain name resolution requests transmitted in accordance with the IP protocol.

4. The method of claim 1, wherein the dynamic initial timeout value is adaptively determined based on the time within which a majority of previous responses are received at the first device over the network in response to historical requests.

5. The method of claim 4, wherein the majority of previous responses comprise about 95% of the previous responses.

6. The method of claim 1, wherein the dynamic initial timeout value is adaptively determined based on the time within which one or more previous responses are received at the first device from the second device.

7. The method of claim 1, wherein the dynamic initial timeout value is based on the RTT time between the first device and the second device.

8. The method of claim 1, wherein the dynamic initial timeout value is based on the time within which one or more previous responses are received at the first device from a third device different from the second device.

9. A system comprising:

a memory storing instructions and data, the data comprising a first request for first information to be provided over a network to a second device;

a processor to execute the instructions and to process the data;

wherein the instructions comprise;

adaptively determining, in an application running on a application layer in a first device, a dynamic initial timeout value with the processor, where the dynamic initial timeout value is based on a time within which one or more previous responses were received at the first device over a network to one or more historical requests, wherein the historical request is a non-pending and closed request;

transmitting the first request for the first information from the first device to the second device over the network;

before a response is received from the second device to the first request, performing a comparison between the time that has elapsed since the first request was transmitted and the adaptively determined dynamic initial timeout value; and

,

transmitting, depending on the comparison and via the network, a new request for the first information to the second device, where the new request is distinct from and unrelated to the first request.

10. The system of claim 9, wherein the first request and the new request are network connection requests transmitted in accordance with the TCP/IP protocol.

11. The system of claim 9, wherein the first request and the new request are domain name resolution requests transmitted in accordance with the IP protocol.

12. The system of claim 9, wherein the dynamic initial timeout value is based on the time within which a majority of previous responses are received at the first device over the network.

13. The system of claim 9, wherein the majority of previous responses comprise about 95% of the responses.

14. The system of claim 9, wherein the dynamic initial timeout value is based on the time within which one or more previous responses are received at the first device from the second device.

15. The system of claim 9, wherein the dynamic initial timeout value is based on the RTT time between the first device and the second device.

16. The system of claim 9, wherein the dynamic initial timeout value is based on the time within which one or more previous responses are received at the first device from a third device different from the second device.