HTTP OPTIMIZATION, MULTI-HOMING, MOBILITY AND PRIORITY

• US 20110222404A1
• Filed: 12/10/2010
• Published: 09/15/2011
• Est. Priority Date: 12/18/2009
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Watson, Mark, Vicisano, Lorenzo
• Granted Patent: US 8,964,757 B2
• Time in Patent Office: Days
• US Class Current: 370/235
• CPC Class Codes: H04L 67/62 Establishing a time schedule for servicing the requests
• CPC Class Codes: H04L 69/14 Multichannel or multilink protocols
• CPC Class Codes: H04W 28/10 Flow control between communication endpoints
• CPC Class Codes: H04W 76/16 Involving different core network technologies, e.g. a packet-switched [PS] bearer in combination with a circuit-switched [CS] bearer
• CPC Class Codes: H04W 8/04 Registration at HLR or HSS [Home Subscriber Server]
• CPC Class Codes: H04W 88/06 adapted for operation in multiple networks or having at least two operational modes, e.g. multi-mode terminals

Abstract

Combining parallel Hypertext Transfer Protocol (HTTP) connections and pipelining overcomes an impact of increasing Round Trip Time (RTT) by varying in real time the number of parallel connections and pipelined requests such that the number of outstanding requests is minimal and the link remains fully utilized. Optimal construction and scheduling of requests and connections in an HTTP stack improves page load time and also provides for greater responsiveness to changes in object priorities. Multi-homing and mobility at the application layer for HTTP are addressed. Multi-homing provides for simultaneous use of multiple interfaces, for example WWAN and WLAN interfaces which improves download time, especially in the case that the available bandwidth the interfaces is of the same order of magnitude. Mobility provides for switching connections as the device moves. In combination they provide for smoother mobility. Mobility can be provided this way without server or network support.

Claims

1. A method for packet data communications, the method comprising:

establishing a number of parallel connections for packet data communication;

transmitting a plurality of pipelined requests via the number of parallel connections for retrieving a hypertext object comprised of packet data portions respectively stored on a plurality of servers; and

dynamically varying the number of parallel connections and pipelined requests via the parallel connections to reduce outstanding requests while maintaining full link utilization.

2. The method of claim 1, wherein transmitting the plurality of pipelined requests further comprises accessing a node of a wireless wide area network.

3. The method of claim 1, wherein dynamically varying the number of parallel connections and pipelined requests further comprises:

estimating bandwidth and round trip time; and

continuing to transmit pipelined requests until a likelihood exists that requested but not received packet data exceeds a product of the estimated bandwidth and round trip time.

4. The method of claim 1, wherein establishing the number of parallel connections for packet data communication further comprises:

minimizing a number of connections to each of the plurality of servers;

using the number of parallel connections to reduce overall congestion response to late or lost requested packet data; and

dynamically varying the number of parallel connections and pipelined requests to optimize conflicting objectives of minimizing the number of connections and using the number of parallel connections to reduce the overall congestion.

5. The method of claim 1, further comprising:

determining a progressive ordering for rendering the packet data portions; and

prioritizing an order of the plurality of pipelined requests in relation to the progressive ordering.

6. The method of claim 1, wherein maintaining full link utilization further comprises utilizing a greedy request opportunity.

7. The method of claim 1, wherein maintaining full link utilization further comprises utilizing a just-in-time request opportunity.

8. The method of claim 1, wherein transmitting the plurality of pipelined requests further comprises requesting header compression.

9. The method of claim 1, further comprising receiving server identified resources that were not requested.

10. The method of claim 9, further comprising:

determining that the server identified resources are not needed; and

canceling the server identified resources.

11. The method of claim 1, further comprising transmitting the plurality of pipelined requests further comprising transmitting a plurality of partial pipelined requests using a header range feature.

12. The method of claim 1, wherein transmitting the plurality of pipelined requests further comprises:

detecting that an interface has become unavailable or has become available;

identifying open connections using the interface;

constructing a list of incomplete pipelined requests that have been issued on the open connections; and

dynamically varying the number of parallel connections and pipelined requests via the parallel connections with connection round robin based in part upon the identified open connections, the interface, and the list of incomplete pipelined requests.

13. The method of claim 1, further comprising maintaining a separate Domain Name Service (DNS) cache for each interface.

14. At least one processor for packet data communications, the at least one processor comprising:

A first module for establishing a number of parallel connections for packet data communication;

a second module for transmitting a plurality of pipelined requests via the number of parallel connections for retrieving a hypertext object comprised of packet data portions respectively stored on a plurality of servers; and

a third module for dynamically varying the number of parallel connections and pipelined requests via the parallel connections to reduce outstanding requests while maintaining full link utilization.

15. A computer program product for packet data communications, the computer program product comprising:

A non-transitory computer-readable medium storing sets of code comprising;

a first set of codes for causing a computer to establish a number of parallel connections for packet data communication;

a second set of codes for causing the computer to transmit a plurality of pipelined requests via the number of parallel connections for retrieving a hypertext object comprised of packet data portions respectively stored on a plurality of servers; and

a third set of codes for causing the computer to dynamically vary the number of parallel connections and pipelined requests via the parallel connections to reduce outstanding requests while maintaining full link utilization.

16. An apparatus for packet data communications, the apparatus comprising:

means for establishing a number of parallel connections for packet data communication;

means for transmitting a plurality of pipelined requests via the number of parallel connections for retrieving a hypertext object comprised of packet data portions respectively stored on a plurality of servers; and

means for dynamically varying the number of parallel connections and pipelined requests via the parallel connections to reduce outstanding requests while maintaining full link utilization.

17. An apparatus for packet data communications, the apparatus comprising:

a transceiver for establishing a number of parallel connections for packet data communication;

the transceiver further for transmitting a plurality of pipelined requests via the number of parallel connections for retrieving a hypertext object comprised of packet data portions respectively stored on a plurality of servers; and

a computing platform for dynamically varying the number of parallel connections and pipelined requests via the parallel connections to reduce outstanding requests while maintaining full link utilization.

18. The apparatus of claim 17, wherein the transceiver is further for transmitting the plurality of pipelined requests by accessing a node of a wireless wide area network.

19. The apparatus of claim 17, wherein the computing platform is further for dynamically varying the number of parallel connections and pipelined requests by:

estimating bandwidth and round trip time; and

continuing to transmit pipelined requests until a likelihood exists that requested but not received packet data exceeds a product of the estimated bandwidth and round trip time.

20. The apparatus of claim 17, wherein the computing platform via the transceiver is further for establishing the number of parallel connections for packet data communication by:

minimizing a number of connections to each of the plurality of servers;

using the number of parallel connections to reduce overall congestion response to late or lost requested packet data; and

dynamically varying the number of parallel connections and pipelined requests to optimize conflicting objectives of minimizing the number of connections and using the number of parallel connections to reduce the overall congestion.

21. The apparatus of claim 17, wherein the computing platform is further for determining a progressive ordering for rendering the packet data portions, and for prioritizing an order of the plurality of pipelined requests in relation to the progressive ordering.

22. The apparatus of claim 17, wherein the computing platform is further for maintaining full link utilization by utilizing a greedy request opportunity.

23. The apparatus of claim 17, wherein the computing platform is further for maintaining full link utilization by utilizing a just-in-time request opportunity.

24. The apparatus of claim 17, wherein the transceiver is further for transmitting the plurality of pipelined requests by requesting header compression.

25. The apparatus of claim 17, wherein the transceiver is further for receiving server identified resources that were not requested.

26. The apparatus of claim 25, wherein the computing platform is further for determining that the server identified resources are not needed, and for canceling the server identified resources.

27. The apparatus of claim 17, wherein the transceiver is further for transmitting the plurality of pipelined requests by transmitting a plurality of partial pipelined requests using a header range feature.

28. The apparatus of claim 17, wherein the computing platform is further for detecting that an interface has become unavailable or has become available;

identifying open connections using the interface;

constructing a list of incomplete pipelined requests that have been issued on the open connections; and

dynamically varying the number of parallel connections and pipelined requests via the parallel connections with connection round robin based in part upon the identified open connections, the interface, and the list of incomplete pipelined requests.

29. The apparatus of claim 17, wherein the computing platform is further for maintaining a separate Domain Name Service (DNS) cache for each interface.