AGGREGATING CONNECTION MAINTENANCE TO OPTIMIZE RESOURCE CONSUMPTION

• US 20120089720A1
• Filed: 12/19/2011
• Published: 04/12/2012
• Est. Priority Date: 03/26/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Tran, Anh P., Moore, Lloyd Alfred, Maabreh, M. Omar, Sapek, Adam, Waliczek, Gary Jason, Sandadi, Upender R.
• Granted Patent: US 8,521,887 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: H04L 41/5019 Ensuring fulfilment of SLA

Abstract

Combining network connection maintenance operations that use a resource to extend battery life. Each of the network connections has a timer and a defined tolerance for early connection maintenance. After receiving notification of an event, the network connections are accessed to identify the connections for which a maintenance operation may be performed early based on the timer and the tolerance. In an embodiment, the maintenance operation includes sending a transmission control protocol (TCP) keep-alive packet on the connection.

Claims

1. A system comprising:

a memory area for storing a timer value, a maximum interval duration, and a tolerance factor for each of a plurality of open transmission control protocol (TCP) connections, said tolerance factor indicating early availability of the associated TCP connection for maintenance operations; and

a processor associated with a computing device, said processor configured to execute computer-executable instructions for;

receiving the maximum interval duration and the tolerance factor for each of the TCP connections;

storing the received maximum interval duration and the received tolerance factor in the memory area;

adjusting the timer value for each of the TCP connections as a function of an elapsed time since activity on the TCP connection;

receiving notification of an event at a notification time;

identifying, in response to the received notification, one or more of the TCP connections to which the received notification applies and for which a difference between the timer value and the maximum interval duration is within the tolerance factor to aggregate the maintenance operations of the identified TCP connections; and

sending a keep-alive packet over each of the identified TCP connections.

2. The system of claim 1, wherein the tolerance factor comprises an argument in an input/output control interface.

3. The system of claim 1, further comprising means for identifying the one or more of the TCP connections as a function of the received event notification, the timer value, and the tolerance factor of each of the TCP connections.

4. The system of claim 1, further comprising means for altering the timer value in a TCP stack for each of the TCP connections to coordinate sending TCP keep-alive packets to minimize use of network resources.

5. A method comprising:

receiving, by a computing device, an event notification;

accessing a plurality of open network connections on the computing device, each of said network connections having associated therewith a timer value and a tolerance factor, said tolerance factor indicating early availability of the associated network connection for maintenance operations;

identifying, by the computing device, one or more of the accessed network connections as a function of the received event notification, the timer value, and the tolerance factor of each of the accessed network connections to aggregate the maintenance operations for the identified network connections in response to the received event notification and availability of the maintenance operations to be performed earlier than scheduled based on the tolerance factor; and

performing, by the computing device, one or more of the maintenance operations on the identified network connections.

6. The method of claim 5, wherein performing the maintenance operation on the identified network connections comprises sending, by a kernel level application executing on the computing device, a keep-alive packet on each of the identified network connections.

7. The method of claim 5, wherein each of the plurality of network connections has an interval duration defining a maximum permissible time of inactivity on the network connection, and further comprising resetting the timer value for each of the identified network connections based on the interval duration for the network connection.

8. The method of claim 7, wherein resetting the timer value comprises adding the interval duration for each of the network connections to a time of receipt of the notification.

9. The method of claim 7, wherein the tolerance factor comprises a percentage value, and wherein identifying the one or more of the accessed network connections comprises:

applying the percentage value to the interval duration for each of the network connections to create a time value;

comparing the timer value to the time value for each of the network connections; and

identifying one or more of the network connections based on said comparing.

10. The method of claim 5, wherein the tolerance factor comprises a time value, and wherein identifying the one or more of the accessed network connections comprises:

comparing the timer value to the time value for each of the network connections; and

identifying one or more of the network connections based on said comparing.

11. The method of claim 5, wherein receiving the event notification comprises receiving notification that a network resource is available, and wherein identifying the one or more of the accessed network connections comprises identifying one or more of the network connections having maintenance operations associated therewith that use the available resource during performance.

12. The method of claim 5, further comprising:

tracking activity on each of the network connections; and

resetting the timer value for each of the network connections responsive to said tracking.

13. The method of claim 5, wherein receiving the event notification comprises receiving notification of a user mode event.

14. The method of claim 5, further comprising receiving the tolerance factor for one of the network connections from an application program associated with said one of the network connections.

15. The method of claim 5, wherein accessing the plurality of open network connections on the computing device comprises accessing a plurality of open hypertext transfer protocol (HTTP) connections on the computing device.

16. The method of claim 5, wherein performing the one or more of the maintenance operations on the identified network connections comprises performing one or more of the maintenance operations on the identified network connections at a transport layer in a multi-layer protocol model.

17. One or more computer-readable storage media having computer-executable components for aggregating network maintenance operations based on availability of a resource in a computing device and on availability of the network maintenance operations to be performed earlier than scheduled based on a tolerance factor, said components comprising:

an interface component for receiving notification of availability of a network resource on the computing device;

a network component for identifying a plurality of open transmission control protocol (TCP) connections that use the network resource, each of said TCP connections having associated therewith a timer and the tolerance factor, said tolerance factor indicating early availability of the associated TCP connection for one or more of the network maintenance operations;

an aggregation component for selecting, in response to the notification of availability of the network resource received by the interface component, one or more of the identified plurality of TCP connections for which time remaining on the timer is within the tolerance factor; and

a maintenance component for sending, via the network resource, a keep-alive packet on the TCP connections selected by the aggregation component to persist the selected TCP connections, said interface component, said network component, said aggregation component, and said maintenance component executing on the computing device.

18. The computer-readable storage media of claim 17, further comprising a timer component for maintaining the timer associated with each of the TCP connections.

19. The computer-readable storage media of claim 18, wherein the timer component maintains the timer with a granularity on the order of seconds.

20. The computer-readable storage media of claim 17, wherein the network resource comprises a transceiver associated with the computing device.