DISTRIBUTED INFORMATION TECHNOLOGY INFRASTRUCTURE DYNAMIC POLICY DRIVEN PEAK POWER MANAGEMENT SYSTEM

• US 20160077570A1
• Filed: 10/28/2014
• Published: 03/17/2016
• Est. Priority Date: 09/12/2014
• Status: Active Grant

• Current Assignee: Vigyanlabs Inc.
• Original Assignee: Vigyanlabs Inc.
• Inventors: Hariharan, Sanjaya Ganesh, Varadarajan, Srinivas, Krishnaswamy, Srivatsa
• Granted Patent: US 9,477,281 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 1/26 Power supply means, e.g. regulation thereof for memories G11C
• CPC Class Codes: G06F 1/32 Means for saving power
• CPC Class Codes: H02J 9/002 in which a reserve is maintained in an energy source by disconnecting non-critical loads, e.g. maintaining a reserve of charge in a vehicle battery for starting an engine
• CPC Class Codes: H04L 12/10 Current supply arrangements
• CPC Class Codes: Y02B 70/30 Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
• CPC Class Codes: Y02D 30/50 in wire-line communication networks, e.g. low power modes or reduced link rate
• CPC Class Codes: Y02P 80/10 Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
• CPC Class Codes: Y04S 20/20 End-user application control systems

Abstract

A peak power management system for networked smart IT devices. These smart devices have computing capability with at least one CPU and memory and can be networked. An uninterruptible power supply provides power to the smart devices. A central intelligent power management server controls the power consumed by all the smart devices networked with the server. The system uses priority based peak power management policies for smart IT devices assisted by fine grain control of external power drawn by each device. By applying different power management policies at different scheduled intervals and controlling the power consumption on the smart devices, the aggregated peak power demand is controlled. The policies can be adapted in-time to suit the actual, real-time power requirement of devices, their priorities, and applicable peak power limit at that time. Also, dynamic policy based peak power management can be extended to an intelligent hierarchical power distribution network.

Claims

1. A peak power management system comprising:

at least one computing device comprising;

a central processing unit (CPU);

a main memory component accessed by said CPU;

a persistent storage connected to said memory component and accessed by said CPU;

a communication component for connecting to a communication network; and

an operating system controlling said at least one computing device and operated by said CPU;

a power management server comprising;

a cluster of computers running operating systems;

a set of virtual machines running on said cluster of computers;

one or more communication components for connecting to a communication network;

a persistent storage for storing any of software programs, applications, and data,

wherein said power management server runs a set of software applications comprising a power management program; and

said communication network that connects said power management server and said at least one computing device for exchanging messages between them,

wherein said power management system draws its electrical power for all said at least one computing device and said power management server from an uninterrupted power supply (UPS),

wherein said UPS draws power from any of an external power supply source and a captive electrical power generator,

wherein said power supply source is configured to impose a temporal peak power profile (TPPP) comprising a time schedule based limits on the amount of peak power that can be drawn by an entity under consideration during various time periods in a year,

wherein said data in the form of said TPPP is stored in said power management server persistently,

wherein said UPS comprises a built-in computing device and is configured to receive commands from said power management program and send responses to said power management program in addition to sending unsolicited messages to said power management program,

wherein each said at least one computing device runs a program comprising an intelligent power management agent all the time independent of whether the agent comprises a connected mode and disconnected mode respectively in order to communicate with said power management program,

wherein said power management program communicates with said agent on all said at least one computing device to send commands and receive responses or receive data from said agents in an unsolicited and time schedule based manner,

wherein said agent controls ae peak power consumption based on an ability to control power consumption of said at least one computing device with minimal impact on a performance of said at least one computing device, and

wherein said agent continues to control power consumption even when said agent is operating in a disconnected mode and is not communicating with said power management program.

2. The peak power management system of claim 1, wherein each said at least one computing device is assigned any of an operational priority and a rank in the range of 1 to 10 indicating a level of importance for said at least one computing device with number 1 being a highest priority and number 10 being a lowest priority, and wherein a first device of higher priority is entitled to a higher external peak power limit than a second device of lower priority, and wherein the peak power limit ranges from a percentage range of 100% to 0% of external power drawn from said UPS.

3. The peak power management system of claim 2, wherein a percentage of peak power for a particular device comprises the percentage of a peak power demand that said particular device can make under unrestricted conditions, wherein all devices of a particular priority are entitled to a same peak power limit at any point of time.

4. The peak power management system of claim 3, further comprising a limited number of minimum peak power allowed (MINPP) data sets corresponding to a peak power limit allowed by a percentage of a set of member devices, wherein a number of said MINPP data sets is equal to a number of priorities plus one.

5. The peak power management system of claim 4, wherein said particular device at any point of time is assigned to be in one of said MINPP data sets by said power management program, wherein an assignment to said MINPP data sets is configured to be changed by said power management program only, and wherein said assignment is restricted by said TPPP.

6. The peak power management system of claim 5, wherein said power management program in said power management server is configured to create a device peak power management policy (DPPP) for each said at least one computing device based on a priority of said at least one computing device, wherein said DPPP comprises an entitlement or allocation of said peak power limit for said at least one computing device at any particular point of time such that a total peak power allocated for all devices does not exceed the peak power limit imposed by said external power supply source as expressed in TPPP at any time.

7. The peak power management system of claim 6, further comprising at least one peak power assignment functions (PPAF) based on said TPPP and operational priorities of all said at least one computing device.

8. The peak power management system of claim 7, wherein said power management program in said power management server sends each DPPP to a corresponding device such that said agent running on said at least one computing device is configured to use said DPPP as a guide to control power consumption at any point of time even while said agent is operating in a disconnected mode.

9. The peak power management system of claim 8, further comprising an enhanced PPAF that further optimizes peak power allocation dynamically by:

considering an actual number of devices that are on and are consuming power;

reviewing a peak power allocation to devices;

calculating a peak power available as a difference between an allocated enterprise peak power (EPP) limit and actually used peak power as measured from said UPS at an end of each review period or cycle;

ensuring that when said available peak power is >

0 then at every review cycle said peak power limit is increased for devices that are entitled for more allocation up from an actual current allocation for a given device, wherein an increased allocation starts with least peak power allocated devices in a descending order of the operational priority of a selected device such that higher priority devices are allocated first among a set of least power allocated devices; and

ensuring that when said available peak power is <

0 such that more power is actually being consumed than the allowed EPP level at that time, then at every review cycle, said peak power limit is decreased for devices from said actual current allocation, wherein a decrease in allocation starts with least peak power allocated devices and among them in an ascending order of operational priority such that a decrease occurs from a lowest priority device to a highest priority device within said MINPP set of devices having a same current peak power allocation.

10. The peak power management system of claim 9, wherein said power management server and all controlled devices have their system clocks synchronized in time using a network time protocol (NTP) thereby ensuring that all devices will look at UPS power supply sinusoidal waves synchronously.

11. The peak power management system of claim 10, wherein in a one second interval, for a 50 Hertz frequency, a complete set of sinusoidal waves cycles are configured to be numbered from 1 to 50 and are seen in a same way by all devices at any time, wherein drawing external power supply from said UPS during all 50 cycles means 100% main peak power allocation and drawing power during only 5 contiguous cycles means 10% peak power allocation.

12. The peak power management system of claim 11, wherein an assignment of a set of contiguous sinusoidal cycles to any device is computed by said power management program based on said DPPP for that device and an overall boundary condition that said EPP is never exceeded at any time by a total allocation for all devices, wherein a random assignment of cycles to all devices is configured to achieve a statistically uniform distribution of power that is limited by said EPP over any given time schedule, and wherein the random assignment of cycles to devices is configured to avoid unexpected total peak power demand exceeding said EPP and is configured to make said total peak power demand flat during a time schedule during which said EPP is applicable.

13. The peak power management system of claim 12, wherein said DPPP comprises several time schedules with each schedule indicating a peak power allocated, a starting sinusoidal wave cycle number, and an ending cycle number for drawing external power from said UPS by a power supply unit of said particular device.

14. The peak power management system of claim 13, wherein said agent invokes programming calls with start and end cycle numbers as input parameters to a power supply programming interface at any of a beginning of each schedule and whenever a new DPPP is sent from said power management program on the server to said at least one computing device, wherein a power supply connects to an external power source only during the cycles to said external power supply and is configured to store power or charge in order to supply an average power to all the internal device components, wherein selective connectivity is achieved by using a digital switching gate to an external source which opens and closes based on contents of a digital counter, and wherein external power charges local batteries on said at least one computing device so that power is supplied continuously to said at least one computing device internally from said local batteries.

15. The peak power management system of claim 6, wherein said power management program collects all power consumption data from said at least one device through said agents and maintains a history database of such data persistently, wherein said peak power management system is configured to allow specific requests for power allocation during specific intervals of any day of a year for specific devices, wherein based on historical data and specific requests, said power management program is able to statistically predict required maximum peak power allocations for any of said at least one computing device for different time intervals of each day and accordingly assign said particular device to a corresponding set MINPP for each such interval.

16. The peak power management system of claim 15, wherein at a beginning of each day, DPPPs are created and sent to said at least one computing device, wherein said DPPPs are configured to be revised periodically during the day with a period equal to a reassignment review interval in order to adjust to real-time peak power demands coming from many devices, wherein said devices that are operating within a predetermined percentage of the allowed peak power in a review interval are configured to send a request to said power management program to allocate more power, and wherein said power management program collects all requests and adaptively reallocates peak power to a subset of requesting devices in order to satisfy the requests whenever feasible.

17. The peak power management system of claim 16, wherein an increased allocation to said requesting devices removes excess allocation from other devices which are utilizing additional peak power than a minimum peak power limit that they are entitled to.

18. The peak power management system of claim 1, wherein said at least one computing device comprises an information technology enabled device.

19. The peak power management system of claim 1, wherein said at least one computing device comprises a display device operatively connected to said CPU.

20. The peak power management system of claim 1, wherein said at least one computing device comprises a power backup storage battery used by said at least one computing device.

21. The peak power management system of claim 1, wherein said peak power management system is configured in a hierarchical power distribution network system comprising:

at least one said peak power management system representing an enterprise consumer; and

a hierarchy network of intelligent power distribution nodes belonging to a power supply company starting with a root level main distribution node,

wherein an enterprise is viewed as an intelligent device making variable peak power demand by its parent node,

wherein each power distribution node is adapted to be dynamic policy driven, peak power management of its child-level nodes, and

wherein each node makes best of use of its allocated peak power during any interval of time and allocates to its child-level nodes using peak power management methods.