Power management of networked devices

• US 8,488,500 B2
• Filed: 05/02/2008
• Issued: 07/16/2013
• Est. Priority Date: 05/02/2008
• Status: Expired due to Fees

• Current Assignee: Dhaani Systems Co.
• Original Assignee: Dhaani Systems Co.
• Inventors: Mukherjee, Shankar, Walrand, Jean, Bambos, Nicholas
• Primary Examiner(s): Sheikh, Ayaz
• Assistant Examiner(s): CHAU, PETER P
• Application Number:
  US12/114,721
• Publication Number:
  US 20090274070A1
• Time in Patent Office: 1,901 Days
• Field of Search: None
• US Class Current: 370/311
• CPC Class Codes: G06F 1/266 Arrangements to supply power to external peripherals either directly from the computer or under computer control, e.g. supply of power through the communication port, computer controlled power-strips
• CPC Class Codes: G06F 1/3234 Power saving characterised by the action undertaken
• CPC Class Codes: G06F 1/3287 by switching off individual functional units in the computer system
• CPC Class Codes: H04W 52/02 Power saving arrangements in wired systems H04L12/12; signaling of mobile application services, e.g. low battery notifications H04W4/20
• CPC Class Codes: H04W 52/0209 in terminal devices terminal devices per se H04W88/02
• CPC Class Codes: H04W 52/0212 managed by the network, e.g. network or access point is master and terminal is slave
• CPC Class Codes: H04W 52/0216 using a pre-established activity schedule, e.g. traffic indication frame
• CPC Class Codes: H04W 52/0254 detecting a user operation or a tactile contact or a motion of the device
• CPC Class Codes: H04W 52/0258 controlling an operation mode according to history or models of usage information, e.g. activity schedule or time of day
• CPC Class Codes: Y02D 30/70 in wireless communication networks

Abstract

A network-centric, power management system and method is disclosed for monitoring and controlling device nodes attached to a network. The monitoring and controlling includes collecting and processing information available on the network about the device nodes and using the collected information to manage power on the device nodes.

Claims

1. A system for managing power consumption of a device node coupled to a communication node of a communication network by way of a communication link, the system comprising:

a control node coupled to the communication node and configured for;

learning an activity pattern associated with the device node, where learning the activity pattern includes monitoring, by way of the communication link, activities undertaken at the device node;

estimating future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based on the estimated future activity of the device node; and

sending the command to the device node, through the communication link, to set the device node into a power management mode associated with the estimated future activity of the device node.

2. The system of claim 1, where the power management mode includes turning the device node on or off.

3. The system of claim 1, where learning the activity pattern includes learning a usage pattern associated with the device node and a user.

4. The system of claim 1, where learning the activity pattern includes learning a user location pattern.

5. The system of claim 1, where the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

6. The system of claim 1, where learning the activity pattern includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

7. The system of claim 1, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

8. The system of claim 1, where the control node estimates a probability of a user being proximate to the device node.

9. The system of claim 1, further comprising:

a repository for storing a power usage profile for the device node.

10. The system of claim 9, further comprising:

a display for displaying the power usage profile selected by a system operator; and

a processing facility configured for using the repository and display for extracting the power usage profile of the device node, which is selected by the system operator.

11. The system of claim 10, where the processing facility initiates an alarm at the device node based on evaluation of power usage associated with the device node or the communication network.

12. The system of claim 1, where the communication node includes a switch, hub or router.

13. The system of claim 1, where the device node is a first device node, the communication link is a first communication link, where the system further includes a second device node coupled to the communication node by way of a second communication link and where the command sent by the control node to the first device node is further based on activity observed at the second device node by way of the second communication link.

14. The system of claim 13, where a user is associated with the first and second device nodes and where the activity observed at the second device node includes whether or not the user is active at the second device node.

15. The system of claim 14, where the command is configured to set the first device node into a reduced power mode when the activity at the second device node indicates that the user is active at the second device node.

16. The system of claim 1, where learning the activity pattern includes monitoring activities undertaken at the device node by means other than the communication link.

17. The system of claim 1, where the command is further based on user activity at another device node coupled to the communication node.

18. A method of managing power consumption of a device node coupled to a communication node of a communication network by way of a communication link, comprising:

learning an activity pattern associated with the device node, where learning the activity pattern includes monitoring, by way of the communication link, activities undertaken at the device node;

estimating future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based on the estimated future activity of the device node; and

sending the command to the device node through the communication link, to set the device node into a power management mode associated with the estimated future activity of the device node.

19. The method of claim 18, where the power management mode includes turning the device node on or off.

20. The method of claim 18, where learning the activity pattern includes learning a usage pattern associated with the device node and a user.

21. The method of claim 18, where learning the activity pattern includes learning a user location pattern.

22. The method of claim 18, where the command is further based upon information other than the learned activity pattern, and where the information includes control rules.

23. The method of claim 18, where learning the activity pattern includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

24. The method of claim 18, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

25. The method of claim 18, further comprising estimating a probability value of a user being proximate to the device node.

26. The method of claim 18, where the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

27. A system for managing a device node, comprising:

a control node coupled to the device node by way of a communication link and configured for;

learning an activity pattern at the device node which includes monitoring, by way of the communication link, activities the device node has undertaken;

estimating, by the control node, future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating, by the control node, a command based on the estimated future activity of the device node; and

sending, by the control node, the command to the device node, through the communication link, to set the device node into a power management mode associated with the estimated future activity of the device node.

28. The system of claim 27, where the system is configured to operate in a data center.

29. The system of claim 27, where the sending of the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

30. The system of claim 27, where the device node is a first device node, where the communication link is a first communication link, where the system further includes a second device node coupled to the control node by way of a second communication link and where the control node is further configured for learning an activity pattern at the second device node at least in part by monitoring, by way of the second communication link, activities undertaken at the second device node, and where the command sent to the first device node is further based upon the learned activity pattern of the second device node.

31. A system for managing power consumption of a first device node coupled to a communication node of a communication network by way of a first communication link, with the system further including a second device node coupled to the communication node by way of a second communication link, the system comprising:

a control node coupled to the communication node and configured;

to learn an activity pattern of the first device node based at least in part on activity observed at the first device node, where learning the activity pattern of the first device node includes monitoring, by way of the first communication link, activities undertaken at the first device node;

to learn an activity pattern of the second device node based at least in part on activity observed at the second device node, where learning the activity pattern of the second device node includes monitoring, by way of the second communication link, activities undertaken at the second device node;

to estimate future activity of the first device node based at least in part on the learned activity patterns of the first and second device nodes, where the estimated future activity of the first device node is based on a probability value that the first device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

to automatically generate a command based on the estimated future activity of the first device node; and

to send the command to the first device node, through the first communication link, to set the first device into a power management mode associated with the estimated future activity of the first device node.

32. The system of claim 31, where a user is associated with the first and second device nodes and the learned activity pattern of the second device node includes whether or not the user is active at the second device node.

33. The system of claim 31, where the command is configured to set the first device node into a lower power mode of reduced power consumption when an additional activity is observed at the second device node indicates that a user is active at the second device node.

34. A system for managing power consumption of a device node coupled to a communication node of a communication network by way of a communication link, with the device node carrying out a first group of tasks on behalf of a first device node user while also carrying out a second group of tasks not associated with the first device node user, the system comprising:

a control node coupled to the communication node and configured for;

learning an activity pattern at the device node at least in part by way of the communication link, where learning the activity pattern includes monitoring, by way of the communication link, activities undertaken at the device node, where the learned activity pattern is based on a first set of operational data reflecting the first group of tasks the device node has undertaken and based on a second set of operational data reflecting the second group of tasks the device node has undertaken;

estimating, by the control node, future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating, by the control node, a command based at least in part on the estimated future activity of the device node; and

sending, by the control node, a command to the device node, through the communication link, with the command being configured to set the device node into power management mode associated with the estimated future activity of the device node.

35. The system of claim 34, where the second group of tasks are carried out on the device node on behalf of a second user and where the estimated future activity further comprises a second probability value that the second user is active on the device node that is derived from the second set of operational data by the control node, with the second probability value being based on the learned activity pattern.

36. A system for managing power consumption of a device node coupled to a communication node of a communication network by way of a communication link, the system comprising:

a control node coupled to the communication node and configured for;

learning a plurality of activity patterns associated with the device node, where learning a plurality of activity patterns includes monitoring, by way of the communication link, activities undertaken at the device node;

estimating future activity of the device node based on at least one learned activity pattern of the plurality of learned activity patterns, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based at least in part on the estimated future activity; and

sending the command to the device node, through the communication link, to set the device node into a power management mode associated at least in part with the estimated future activity of the device node.

37. The system of claim 36, where the power management mode includes turning the device node on or off.

38. The system of claim 36, where learning the plurality of activity patterns includes learning a usage pattern associated with the device node and a user.

39. The system of claim 36, where learning the plurality of activity patterns includes learning a user location pattern.

40. The system of claim 36, where the sending of the command to the device node includes sending the command in response to the control node determining that a user has gone inactive.

41. The system of claim 36, where learning the plurality of activity patterns includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

42. The system of claim 36, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

43. The system of claim 36, where the control node estimates a probability of a user being proximate to the device node.

44. The system of claim 36, further comprising:

a repository for storing a power usage profile for the device node.

45. The system of claim 44, further comprising:

a display for displaying a power usage profile; and

a processing facility for using the repository and display for extracting the power usage profile of the device node, which is selected by a system operator.

46. The system of claim 45, where the processing facility initiates an alarm at the device node based on evaluation of power usage associated with the monitored device node or the communication network.

47. The system of claim 36, where the communication node includes a switch, hub or router.

48. The system of claim 36, where the system is configured to operate in a data center.

49. The system of claim 36, where the device node is a first device node, the communication link is a first communication link, where the system further includes a second device node coupled to the communication node by way of a second communication link and where the command sent by the control node to the first device node is further based on activity observed at the second device node by way of the second communication link.

50. The system of claim 49, where a user is associated with the first and second device nodes and where the activity observed at the second device node includes whether or not the user is active at the second device node.

51. The system of claim 36, where the device node is a first device node, where the communication link is a first communication link, where the system further includes a second device node coupled to the communication node by way of a second communication link and where the control node is further configured for learning an activity pattern at the second device node at least in part by monitoring, by way of the second communication link, activities undertaken at the second device node, and where the command sent to the first device node is further based upon the learned activity pattern of the second device node.

52. A system for managing power consumption of a device node coupled to a communication node of a communication network by way of a communication link, with the device node carrying out a first group of tasks on behalf of a first device node user while also carrying out a second group of tasks not associated with the first device node user, the system comprising:

a control node coupled to the communication node and configured for;

learning a plurality of activity patterns at the device node at least in part by way of the communication link, where learning the plurality of activity patterns includes monitoring, by way of the communication link, activities undertaken at the device node, where the learned plurality of activity patterns are based on a first set of operational data reflecting the first group of tasks the device node has undertaken and a second set of operational data reflecting the second group of tasks the device node has undertaken;

estimating, by the control node, future activity of the device node based at least in part on the learned plurality of activity patterns, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating, by the control node, a command based at least in part on the estimated future activity of the device node; and

sending, by the control node, the command to the device node, through the communication link, with the command being configured to set the device node into a power management mode associated with the estimated future activity of the device node.

53. The system of claim 52, where the second group of tasks are carried out on the device node on behalf of a second user and a second probability value that the second user is active on the device node is derived from the second set of operational data by the control node, with the second probability value being based on the learned plurality of activity patterns.

54. A non-transitory computer-readable medium having instructions stored thereon, which, when executed by one or more processors in a control node coupled to a communication node of a communication network, causes the one or more processors to perform operations comprising:

learning an activity pattern associated with a device node coupled to the communication node by way of a communication link, where learning the activity pattern includes monitoring, by way of the communication link, activities undertaken at the device node;

estimating future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based on the estimated future activity of the device node; and

sending the command to the device node, through a communication link, to set the device node into a power management mode associated with the estimated future activity of the device node.

55. The non-transitory computer-readable medium of claim 54, where the power management mode includes turning the device node on or off.

56. The non-transitory computer-readable medium of claim 54, where learning the activity pattern includes learning a usage pattern associated with the device node and a user.

57. The non-transitory computer-readable medium of claim 54, where learning the activity pattern includes learning a user location pattern.

58. The non-transitory computer-readable medium of claim 54, where the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

59. The non-transitory computer-readable medium of claim 54, where learning the activity pattern includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

60. The non-transitory computer-readable medium of claim 54, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

61. The non-transitory computer-readable medium of claim 54, where the control node estimates a probability of a user being proximate to the device node.

62. The non-transitory computer-readable medium of claim 54, further comprising instructions for maintaining a repository for storing a power usage profile of the device node.

63. The non-transitory computer-readable medium of claim 62, further comprising:

instructions for presenting on a display the power usage profile when selected by a system operator; and

instructions for extracting from the repository, the power usage profile of the device node when selected by the system operator.

64. The non-transitory computer-readable medium of claim 63, where the instructions activate an alarm at the device node based on evaluation of power usage associated with the device node or the communication network.

65. The non-transitory computer-readable medium of claim 54, where the command is further based upon information other than the learned activity pattern, and where the information includes control rules.

66. The non-transitory computer-readable medium of claim 54, where the one or more processors, which execute the instructions, are in a data center.

67. The non-transitory computer-readable medium of claim 54, where learning the activity pattern includes monitoring activities undertaken at the device node by means other than the communication link.

68. The non-transitory computer-readable medium of claim 54, where the command is further based on user activity at another device node coupled to the communication node.

69. The non-transitory computer-readable medium of claim 54, where the one or more processors in the control node install and cause to execute in the device node instructions responsible for learning the activity pattern associated with the device node.

70. The non-transitory computer-readable medium of claim 69, where the one or more processors in the control node further install and cause to execute in the device node instructions responsible for estimating future activity of the device node based at least in part on the learned activity pattern.

71. A non-transitory computer-readable medium having instructions stored thereon, which, when executed by one or more processors in a control node coupled to a communication node of a communication network, causes the one or more processors to perform operations comprising:

learning an activity pattern at a first device node coupled to the communication node by way of a first communication link, where the activity pattern is based at least in part on activity observed at the first device node, and where learning the activity pattern of the first device node includes monitoring, by way of the first communication link, activities undertaken at the first device node;

learning an activity pattern at a second device node coupled to the communication node of the communication network by way of a second communication link based at least in part on activity observed at the second device node, where learning the activity pattern of the second device node includes monitoring, by way of the second communication link, activities undertaken at the second device node; and

estimating future activity of the first device node based at least in part on the learned activity patterns of the first and second device nodes, where the estimated future activity of the first device node is based on a probability value that the first device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based on the estimated future activity of the first device node; and

sending the command to the first device node, through the first communication link, to set the first device node into a power management mode associated with the estimated future activity of the first device node.

72. The non-transitory computer-readable medium of claim 71, where the activity observed at the second device node includes whether or not a user is active at the second device node.

73. The non-transitory computer-readable medium of claim 71, where the command is configured to set the first device node into a reduced power mode when the activity observed at the second device node indicates that a user is active at the second device node.

74. The non-transitory computer-readable medium of claim 71, where the command sent by the control node to the first device node is further based on activity observed at the second device node by way of a second communication link.

75. The non-transitory computer-readable medium of claim 71, where the power management mode includes turning the first device node on or off.

76. The non-transitory computer-readable medium of claim 71, where learning the activity pattern includes learning a usage pattern associated with the first device node and a user.

77. The non-transitory computer-readable medium of claim 71, where learning the activity pattern includes learning a user location pattern.

78. The non-transitory computer-readable medium of claim 71, where the command to the first device node for setting the first device node into a power management mode associated with the estimated future activity of the first device node is sent in response to the control node determining that a user has gone inactive.

79. The non-transitory computer-readable medium of claim 71, where learning the activity pattern includes monitoring one or more of:

user calendar information associated with the first or second device node, user login history associated with the first or second device node, user usage history associated with the first or second device node and user input entered at the first or second device node.

80. The non-transitory computer-readable medium of claim 71, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

81. The non-transitory computer-readable medium of claim 71, where the control node estimates a probability of a user being proximate to the first device node.

82. The non-transitory computer-readable medium of claim 71, further comprising instructions for maintaining a repository for storing a power usage profile of the first device node and a power usage profile of the second device node.

83. The non-transitory computer-readable medium of claim 82, further comprising:

instructions for presenting on a display the power usage profile of the first device when selected by a system operator; and

instructions for extracting from the repository, the power usage profile of the first device node when selected by the system operator.

84. The non-transitory computer-readable medium of claim 83, where the instructions activate an alarm at the first device node based on evaluation of power usage associated with the first device node or the communication network.

85. The non-transitory computer-readable medium of claim 71, where the command is further based upon information other than the learned activity pattern, and where the information includes control rules.

86. The non-transitory computer-readable medium of claim 71, where the one or more processors, which execute the instructions, are in a data center.

87. The non-transitory computer-readable medium of claim 71, where learning the activity pattern includes monitoring activities undertaken at the first device node by means other than the first communication link.

88. The non-transitory computer-readable medium of claim 71, where the one or more processors in the control node install and cause to execute in the first device node instructions responsible for learning the activity pattern associated with the first device node.

89. The non-transitory computer-readable medium of claim 88, where the one or more processors in the control node further install and cause to execute in the first device node instructions responsible for estimating future activity of the first device node based at least in part on the learned activity pattern.

90. A non-transitory computer-readable medium having instructions stored thereon, which, when executed by one or more processors in a control node coupled to a communication node of a communication network, causes the one or more processors to perform operations comprising:

learning an activity pattern at a device node at least in part by way of a communication link, where learning the activity pattern includes monitoring, by way of the communication link, activities undertaken at the device node, where the learned activity pattern is based on a first set of operational data reflecting a first group of tasks the device node has undertaken and a second set of operational data reflecting a second group of tasks the device node has undertaken;

estimating future activity of the device node based at least in part on the learned activity pattern, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based at least in part on the estimated future activity of the device node, where the command is configured to set the device node into a power management mode associated with the estimated future activity of the device node; and

sending the command to the device node through the communication link.

91. The non-transitory computer-readable medium of claim 90, where the second group of tasks are carried out on the device node on behalf of a second user and where the estimated future activity further comprises a second probability value that the second user is active on the device node that is derived from the second set of operational data by the control node, with the second probability value being based on the learned activity pattern.

92. The non-transitory computer-readable medium of claim 90, where the power management mode includes turning the device node on or off.

93. The non-transitory computer-readable medium of claim 90, where learning the activity pattern includes learning a usage pattern associated with the device node and a user.

94. The non-transitory computer-readable medium of claim 90, where learning the activity pattern includes learning a user location pattern.

95. The non-transitory computer-readable medium of claim 90, where the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

96. The non-transitory computer-readable medium of claim 90, where learning the activity pattern includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

97. The non-transitory computer-readable medium of claim 90, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

98. The non-transitory computer-readable medium of claim 90, where the control node estimates a probability of a user being proximate to the device node.

99. The non-transitory computer-readable medium of claim 90, further comprising instructions for maintaining a repository for storing a power usage profile of the device node.

100. The non-transitory computer-readable medium of claim 99, further comprising:

instructions for presenting on a display the power usage profile when selected by a system operator; and

instructions for extracting from the repository, the power usage profile of the device node when selected by the system operator.

101. The non-transitory computer-readable medium of claim 100, where the instructions activate an alarm at the device node based on evaluation of power usage associated with the device node or the communication network.

102. The non-transitory computer-readable medium of claim 90, where the command is further based upon information other than the learned activity pattern, and where the information includes control rules.

103. The non-transitory computer-readable medium of claim 90, where the one or more processors, which execute the instructions, are in a data center.

104. The non-transitory computer-readable medium of claim 90, where learning the activity pattern includes monitoring activities undertaken at the device node by means other than the communication link.

105. The non-transitory computer-readable medium of claim 90, where the command is further based on user activity at another device node coupled to the communication node.

106. The non-transitory computer-readable medium of claim 90, where the one or more processors in the control node install and cause to execute in the device node instructions responsible for learning the activity pattern associated with the device node.

107. The non-transitory computer-readable medium of claim 106, where the one or more processors in the control node further install and cause to execute in the device node instructions responsible for estimating future activity of the device node based at least in part on the learned activity pattern.

124. The non-transitory computer-readable medium of claim 105, where the one or more processors in the control node install and cause to execute in the device node instructions responsible for learning the plurality of activity patterns associated with the device node.

125. The non-transitory computer-readable medium of claim 124, where the one or more processors in the control node further install and cause to execute in the device node instructions responsible for estimating future activity of the device node based at least in part on the learned the plurality of activity patterns.

108. A non-transitory computer-readable medium having instructions stored thereon, which, when executed by one or more processors in a control node coupled to a communication node of a communication network, causes the one or more processors to perform operations comprising:

learning a plurality of activity patterns at a device node coupled to the communication node at least in part by way of a communication link, where learning the plurality of activity patterns includes monitoring, by way of the communication link, activities undertaken at the device node, where the learned plurality of activity patterns are based on a first set of operational data reflecting a first group of tasks the device node has undertaken and a second set of operational data reflecting a second group of tasks the device node has undertaken;

estimating future activity of the device node based at least in part on the learned plurality of activity patterns, where the estimated future activity of the device node is based on a probability value that the device node will be active at a future time and wherein the probability value is calculated by one or more linear filters or non-linear filters;

automatically generating a command based at least in part on the estimated future activity of the device node; and

sending the command to the device node through the communication link, to set the device node into a power management mode associated with the estimated future activity of the device node.

109. The non-transitory computer-readable medium of claim 108, where the learning of the plurality of activity patterns includes monitoring one or more of:

user calendar information associated with the device node, user login history associated with the device node, user usage history associated with the device node and user input entered at the device node.

110. The non-transitory computer-readable medium of claim 108, where the second group of tasks are carried out on the device node on behalf of a second user and where the estimated future activity further comprises a second probability value that the second user is active on the device node that is derived from the second set of operational data by the control node, with the second probability value being based on the learned plurality of activity patterns.

111. The non-transitory computer-readable medium of claim 108, where the power management mode includes turning the device node on or off.

112. The non-transitory computer-readable medium of claim 108, where learning the plurality of activity patterns includes learning a usage pattern associated with the device node and a user.

113. The non-transitory computer-readable medium of claim 108, where learning the plurality of activity patterns includes learning a user location pattern.

114. The non-transitory computer-readable medium of claim 108, where the command to the device node for setting the device node into a power management mode associated with the estimated future activity of the device node is sent in response to the control node determining that a user has gone inactive.

115. The non-transitory computer-readable medium of claim 108, where the estimated future activity is based on one or more of:

a user response to a system request or communication and a sensed activity in a device node environment.

116. The non-transitory computer-readable medium of claim 108, where the control node estimates a probability of a user being proximate to the device node.

117. The non-transitory computer-readable medium of claim 108, further comprising instructions for maintaining a repository for storing a power usage profile of the device node.

118. The non-transitory computer-readable medium of claim 117, further comprising:

instructions for presenting on a display the power usage profile when selected by a system operator; and

instructions for extracting from the repository, the power usage profile of the device node when selected by the system operator.

119. The non-transitory computer-readable medium of claim 118, where the instructions activate an alarm at the device node based on evaluation of power usage associated with the device node or the communication network.

120. The non-transitory computer-readable medium of claim 108, where the command is further based upon information other than the learned the plurality of activity patterns, and where the information includes control rules.

121. The non-transitory computer-readable medium of claim 108, where the one or more processors, which execute the instructions, are in a data center.

122. The non-transitory computer-readable medium of claim 108, where learning the plurality of activity patterns includes monitoring activities undertaken at the device node by means other than the communication link.

123. The non-transitory computer-readable medium of claim 108, where the command is further based on user activity at another device node coupled to the communication node.