System and method for activation of a plurality of servers in dependence on workload trend

• US 9,715,264 B2
• Filed: 06/27/2016
• Issued: 07/25/2017
• Est. Priority Date: 07/21/2009
• Status: Active Grant

• Current Assignee: The Research Foundation for The State University of New York (State University of New York)
• Original Assignee: The Research Foundation for The State University of New York (State University of New York)
• Inventors: Ghose, Kanad
• Primary Examiner(s): PHAN, RAYMOND NGAN
• Application Number:
  US15/193,901
• Publication Number:
  US 20160306410A1
• Time in Patent Office: 393 Days
• Field of Search: 713300-340, 709201-203, 709227-229
• CPC Class Codes: A61K 36/185 Magnoliopsida (dicotyledons)
• CPC Class Codes: B03B 1/00 Conditioning for facilitating separation by altering physical properties of the matter to be treated
• CPC Class Codes: B03B 11/00 Feed or discharge devices integral with washing or wet-separating equipment
• CPC Class Codes: B03B 5/02 using shaken, pulsated or stirred beds as the principal means of separation B03B5/28, B03B5/48 take precedence
• CPC Class Codes: B03B 5/58 Bowl classifiers
• CPC Class Codes: G05D 23/19 characterised by the use of electric means G05D23/1393 takes precedence
• CPC Class Codes: G06F 1/20 Cooling means
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: G06F 1/3206 Monitoring of events, devices or parameters that trigger a change in power modality
• 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
• CPC Class Codes: G06F 13/409 Mechanical coupling back panels H05K7/1438
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/4893 taking into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: G06F 9/5094 where the allocation takes into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: G06K 19/0723 the record carrier comprising an arrangement for non-contact communication, e.g. wireless communication circuits on transponder cards, non-contact smart cards or RFIDs
• CPC Class Codes: G06K 19/07705 the visual interface being a single light or small number of lights capable of being switched on or off, e.g. a series of LEDs
• CPC Class Codes: G06K 19/07722 the record carrier being multilayered, e.g. laminated sheets flat articles in general, see B32B37/00
• CPC Class Codes: H04L 69/329 : in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 : Network security protocols
• CPC Class Codes: H05K 7/20836 : Thermal management, e.g. server temperature control
• CPC Class Codes: Y02D 10/00 : Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A system and method of scheduling tasks, comprising receiving activity and performance data from registers or storage locations maintained by hardware and an operating system; storing calibration coefficients associated with the activity and performance data; computing an energy dissipation rate based on at least the activity and performance data; and scheduling tasks under the operating system based on the computed energy dissipation rate.

Claims

1. A system for controlling a plurality of servers, each server being implemented though at least one automated processor, each respective server having at least an active mode supplying processing capacity for handling a time-varying workload, and an inactive mode unavailable for supplying additional processing capacity for handling the time-varying workload, wherein the plurality of servers in the active mode have a first aggregate processing capacity during a first time period sufficient for handling the time-varying workload, the system for controlling comprising:

at least one automated controller configured to control the plurality of servers, configured to;

determine a parameter pertaining to at least one of a time-varying state of at least one of the plurality of servers in the active mode, and an environment of operation of at least one of the plurality of servers;

determine a workload trend for the system;

determine, based on at least the parameter and the workload trend, a required second aggregate processing capacity for handling the time-varying workload during a second time period, different from the first processing capacity during the first time period, wherein the second time period extends beyond the first time period; and

generate an output based on the required second aggregate processing capacity.

2. The system according to claim 1, wherein the parameter comprises a feedback signal representing at least one current measured time-varying state of at least one of the plurality of servers in the active mode.

3. The system according to claim 1, wherein the parameter comprises a feedback signal representing at least one current measured time-varying state of an environment of operation of at least one of the plurality of servers in the active mode.

4. The system according to claim 1, wherein the required second aggregate processing capacity different from the first processing capacity is determined based on at least the workload trend, and a current state of at least one of the plurality of servers in the active mode.

5. The system of claim 1, wherein the at least one automated controller is further configured to establish the second aggregate processing capacity responsive to an optimization of energy consumption while achieving a desired workload processing performance.

6. The system of claim 1, wherein the at least one automated controller is further configured to select the second aggregate processing capacity, responsive to at least one of a latency for server activation, and a latency for server deactivation.

7. The system of claim 1, wherein the time-varying state comprises at least one of an I/O device utilization parameter, a CPU utilization parameter, a temperature, a job dispatch rate, a job completion rate, and an incoming job queue size.

8. The system of claim 5, further comprising a load distribution switch comprising at least one load balancer configured to distribute incoming time-varying workload among the servers in the active mode in a data center, after changing the plurality of servers in the active mode to correspond to the second aggregate processing capacity.

9. The system of claim 5, wherein the at least one automated controller is further configured:

to select the second aggregate processing capacity predicted to maintain a desired range of at least one predetermined metric for processing time-varying workload during the second time period, and

to produce a signal adapted to transition at least one server between the active mode and the inactive mode, or between the inactive mode and the active mode.

10. The system of claim 5, wherein the at least one automated controller is further configured:

to select the second aggregate processing capacity predicted to maintain a desired range of at least one predetermined metric for processing time-varying workload during the second time period, and

to produce a signal adapted to change a performance capacity of at least one server.

11. A method for controlling a plurality of servers, each respective server being implemented though at least one automated processor, and having at least an active mode supplying processing capacity for handling a time-varying workload, and an inactive mode unavailable for supplying additional processing capacity for handling the time-varying workload, wherein the plurality of servers in the active mode have a first aggregate processing capacity during a first time period, the method comprising:

determining a parameter pertaining to at least one of a time varying state of at least one of the plurality of servers in the active mode, and an environment of operation of at least one of the plurality of servers;

determining a workload trend for the system;

determining, using at least one automated processor, based on at least the parameter and the workload trend, a required second aggregate processing capacity for handling the time-varying workload during a second time period, different from the first processing capacity during the first time period, wherein the second time period extends beyond the first time period; and

communicating a signal responsive to the second aggregate processing capacity.

12. The method according to claim 11, wherein the plurality of servers are in a server pool, further comprising:

determining a deviation of the first aggregate processing capacity of the plurality of servers from a desired value of the second aggregate processing capacity; and

modifying the plurality of servers to obtain the required second aggregate processing capacity during the second time period by changing at least one of an activation mode of at least one server, a processing capacity of at least one server, and a cooling parameter for at least one server.

13. The method according to claim 11, wherein the parameter comprises at least one feedback signal dependent on at least one measured data value pertaining to at least one of the plurality of servers in the active mode.

14. The method according to claim 11, wherein the parameter comprises at least one feedback signal representing an environment of operation of at least one of the plurality of servers in the active mode.

15. The method according to claim 11, further comprising determining the required second aggregate processing capacity different from the first processing capacity based on at least the workload trend, a time-varying state of at least one of the plurality of servers in the active mode, and an environment of operation of at least one of the plurality of servers.

16. The method according to claim 11, further comprising establishing the second aggregate processing capacity responsive to an optimization of energy consumption while achieving a desired workload processing performance.

17. The method according to claim 11, further comprising selecting the second aggregate processing capacity which is different from the first aggregate processing capacity responsive to at least one of a latency for server activation, and a latency for server deactivation.

18. The method of claim 11, wherein the time-varying state comprises at least one of an I/O device utilization parameter, a CPU utilization parameter, a temperature, and a job dispatch rate, a job completion rate, and an incoming job queue size.

19. The method according to claim 16, further comprising load balancing incoming time-varying workload among the servers in the active mode in a data center during the second time period, after changing the plurality of servers in the active mode to correspond to the second aggregate processing capacity.

20. The method according to claim 16, further comprising:

selecting the second aggregate processing capacity predicted to maintain a desired range of at least one predetermined metric for processing time-varying workload during the second time period; and

changing a number of the plurality of servers in the active mode corresponding to the second aggregate processing capacity for processing the time-varying workload.

21. The method according to claim 16, further comprising:

selecting the second aggregate processing capacity predicted to maintain a desired range of at least one predetermined metric for processing time-varying workloads during the second time period, and

changing a performance capacity of at least one server, to achieve the second aggregate processing capacity for processing the time-varying workload.

22. The method according to claim 16, further comprising:

selecting the second aggregate processing capacity predicted to maintain a desired range of at least one predetermined metric for processing time-varying workloads during the second time period;

changing an activation mode of at least one server of the plurality of servers between the active mode and the inactive mode; and

changing a performance capacity of at least one server in the active mode.

to thereby achieve the second aggregate processing capacity for processing the time-varying workload during the second time period.

23. A method for controlling a plurality of servers, each respective server having at least an active mode supplying processing capacity for handling a time-varying workload, and an inactive mode unavailable for supplying additional processing capacity for handling the time-varying workload, the method comprising:

determining a current value of an time-varying state of at least the plurality of servers in the active mode, wherein the plurality of servers in the active mode have a first aggregate processing capacity during a first time period;

determining a workload trend for the system;

determining, based on at least the workload trend and the current value of the time-varying state of at least the plurality of servers in the active mode during the first time period, a required second aggregate processing capacity different from the first processing capacity for handling the time-varying workload during a second time period, wherein the second time period extends beyond the first time period; and

changing the plurality of servers to obtain the second processing capacity, according to an energy consumption optimization of the plurality of servers while achieving a desired workload processing performance during the second time period.