Decentralized Sleep Management

• US 20120117404A1
• Filed: 11/05/2010
• Published: 05/10/2012
• Est. Priority Date: 11/05/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Lorch, Jacob R., Sen, Siddhartha, Padhye, Jitendra D., Hughes, Richard L., Garcia Jurado Suarez, Carlos
• Granted Patent: US 9,582,062 B2
• Time in Patent Office: Days
• US Class Current: 713/323
• CPC Class Codes: G06F 1/3209 Monitoring remote activity, e.g. over telephone lines or network connections
• CPC Class Codes: G06F 1/3228 Monitoring task completion, e.g. by use of idle timers, stop commands or wait commands

Abstract

Techniques for employing a decentralized sleep management service are described herein. In some instances, each computing device of a group of computing devices periodically shares information about itself with each other computing device of the group. With this information, each computing device within the group that is awake and capable of managing other devices selects a subset of devices to probe. The devices then probe this subset to determine whether the probed devices are asleep. In response to identifying a sleeping device, the probing device takes over management of the sleeping device. Managing the sleeping device involves informing other devices of the group that the sleeping device is being managed, in addition to monitoring requests for services on the sleeping device. In response to receiving a valid request for a service hosted by the sleeping device, the managing device awakens the sleeping device and ceases managing the now-woken device.

Claims

1. A computing device residing within a subnet that includes multiple other computing devices, the computing device comprising:

one or more processors; and

one or more computer-readable media storing computer-executable instructions that, when executed, cause the one or more processors to perform acts comprising;

receiving a local state of each of the multiple other computing devices that reside within the subnet;

maintaining a global state of the subnet with use of the received local states;

identifying, from the global state, a computing device of the subnet to probe;

probing the identified computing device to determine whether the identified computing device is asleep and unmanaged; and

managing the identified computing device at least partly in response to determining that that the identified computing device is asleep and unmanaged.

2. A computing device as recited in claim 1, wherein the one or more computer-readable media further store computer-executable instructions that, when executed, cause the one or more processors to perform an act comprising sending a local state of the computing device to each of the multiple other computing devices that reside within the subnet.

3. A computing device as recited in claim 1, wherein the global state of the subnet indicates computing devices that are actively probing and computing devices that are eligible to be probed, and the identifying of the computing device to probe comprises:

selecting a subset of the computing devices that are eligible to be probed, the selected subset having a size that is based at least in part on a number the computing devices that are actively probing; and

probing each computing device of the selected subset.

4. A computing device as recited in claim 1, wherein the managing of the identified computing device includes sending a local state of the identified computing device to each of the multiple other computing devices that reside within the subnet.

5. A computing device as recited in claim 1, wherein the subnet includes one or more switches to route traffic to computing devices within the subnet, and the managing of the identified computing device comprises informing the switch that the computing device is now responsible for traffic intended for the identified computing device.

6. A computing device as recited in claim 1, wherein the global state indicates a set of one or more services hosted by the identified computing device, and the managing of the identified computing device comprises:

receiving a request to access a service hosted by the identified computing device;

determining whether the set of services indicated by the global state includes the service specified in the received request;

waking the identified computing device at least partly in response to determining that the set of services indicated by the global state includes the service specified in the received request; and

refraining from waking the identified computing device at least partly in response to determining that the set of services indicated by the global state does not include the service specified in the received request.

7. A computing device as recited in claim 1, wherein the managing of the identified computing device comprises:

receiving probe requests intended for the identified computing device and sent by other ones of the multiple other computing devices that reside within the subnet; and

at least partly in response to the receiving of the probe requests, informing the other ones of the multiple other computing devices that the identified computing device is currently being managed.

8. A computing device as recited in claim 1, wherein the managing of the identified computing device comprises:

monitoring outbound service requests sent by the computing device that manages the identified computing device;

determining, for each of the outbound service requests, whether the respective outbound service request is for a service hosted by the identified computing device; and

waking the identified computing device at least partly in response to determining that the respective outbound service request is for a service hosted by the identified computing device.

9. A computing device as recited in claim 1, wherein the managing of the identified computing device comprises:

listening for one or more predetermined communications from the identified computing device; and

ceasing the managing of the identified computing device at least partly in response to receiving one of the one or more predetermined communications.

10. A computing device as recited in claim 1, wherein the managing of the identified computing device comprises attempting to wake the identified computing device at a predetermined time.

11. A computing device as recited in claim 1, wherein the one or more computer-readable media further store computer-executable instructions that, when executed, cause the one or more processors to perform acts comprising:

computing a minimum number of management-capable computing devices within the subnet to keep awake at a given time;

determining a number of management-capable computing devices within the subnet that are awake at the given time; and

waking one or more management-capable computing devices within the subnet when the determined number is less than the computed minimum number.

12. A method implemented at least in part by a node of a decentralized wakeup service, the method comprising:

receiving information regarding multiple other nodes of the decentralized wakeup service;

determining that at least one of the multiple other nodes is asleep based at least in part on the received information; and

managing the node that is determined to be asleep at least partly in response to the determining.

13. A method as recited in claim 12, wherein each node of the decentralized wakeup service resides within a common subnet.

14. A method as recited in claim 12, further comprising sharing information regarding the node of the decentralized wakeup service at least partly in response to a change in the information or after expiration of a predetermined amount of time.

15. A method as recited in claim 12, further comprising:

determining that another node of the decentralized wakeup service is managing or attempting to manage the node that is determined to be asleep;

computing, for both the node and the another node, a priority to manage the node that is determined to be asleep;

comparing the computed priority for the node to the computed priority for the another node to determine which node has a greater priority;

continuing the managing of the node that is determined to be asleep at least partly in response to determining that the node has the greater priority; and

ceasing the managing of the node that is determined to be asleep at least partly in response to determining that the another node has the greater priority.

16. A method as recited in claim 12, wherein the managing of the node that is determined to be asleep comprises:

receiving, at the node of the decentralized wakeup service, a service request intended for the node that is determined to be asleep;

determining whether the service request is associated with a service hosted by the node that is determined to be asleep; and

waking the node that is determined to be asleep and ceasing the managing at least partly in response to determining that the service request is associated with a service hosted by the node that is determined to be asleep.

17. A method as recited in claim 16, wherein the managing of the node that is determined to be asleep further comprises:

sending a state of the node that is determined to be asleep to each other node of the decentralized wakeup service;

informing a switch in the decentralized wakeup service that the node is now responsible for traffic intended for the node that is determined to be asleep; and

after the informing of the switch;

receiving requests for an address of the node that is determined to be asleep and replying to the requests on behalf of the node that is determined to be asleep;

receiving probe requests intended for the node that is determined to be asleep and replying to the probe requests on behalf of the node that is determined to be asleep;

monitoring outbound service requests and waking the node that is determined to be asleep for service requests intended for the node that is determined to be asleep;

attempting to wake the node that is determined to be asleep a predetermined amount of time after beginning the managing; and

ceasing the managing of the node that is determined to be asleep at least partly in response to receiving a predetermined communication from the node that is determined to be asleep.

18. One or more computer-readable media storing computer-executable instructions that, when executed, cause one or more processors to perform acts comprising:

receiving state information from each of multiple different computing devices of a decentralized service;

detecting when at least one of the multiple different computing devices of the decentralized service goes to sleep with use of the received state information; and

assuming responsibility for determining when to awaken the at least one computing device.

19. One or more computer-readable media as recited in claim 18, wherein the at least one computing device includes a watch timer that is set to cause the at least one computing device to periodically awaken.

20. One or more computer-readable media as recited in claim 18, further storing computer-executable instructions that, when executed, cause the one or more processors to perform an act comprising awakening the at least one computing device at least partly in response to:

receiving a request for a service hosted by the at least one computing device;

receiving a request for data hosted by the at least one computing device;

or

determining that use of the at least one computing device will aid balancing of a load handled by the decentralized service.