Energy efficiency aware thermal management in a multi-processor system on a chip
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G05B-015/02
G05D-023/19
G06F-001/20
G06F-001/32
G06F-001/26
G06F-011/30
G06F-011/34
출원번호
US-0280630
(2014-05-18)
등록번호
US-9977439
(2018-05-22)
발명자
/ 주소
Park, Hee Jun
Kang, Young Hoon
Alton, Ronald Frank
Medrano, Christopher Lee
Anderson, Jon James
출원인 / 주소
QUALCOMM Incorporated
대리인 / 주소
Smith Tempel Blaha LLC
인용정보
피인용 횟수 :
0인용 특허 :
23
초록▼
Various embodiments of methods and systems for energy efficiency aware thermal management in a portable computing device that contains a heterogeneous, multi-processor system on a chip (“SoC”) are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit
Various embodiments of methods and systems for energy efficiency aware thermal management in a portable computing device that contains a heterogeneous, multi-processor system on a chip (“SoC”) are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit different processing efficiencies at a given temperature, energy efficiency aware thermal management techniques that compare performance data of the individual processing components at their measured operating temperatures can be leveraged to optimize quality of service (“QoS”) by adjusting the power supplies to, reallocating workloads away from, or transitioning the power mode of, the least energy efficient processing components. In these ways, embodiments of the solution optimize the average amount of power consumed across the SoC to process a MIPS of workload.
대표청구항▼
1. A method for managing thermal energy generation in a portable computing device having a synchronous multi-processor system on a chip (“SoC”), the method comprising: monitoring temperature readings uniquely associated with each of a plurality of individual processing components in the multi-proces
1. A method for managing thermal energy generation in a portable computing device having a synchronous multi-processor system on a chip (“SoC”), the method comprising: monitoring temperature readings uniquely associated with each of a plurality of individual processing components in the multi-processor SoC, wherein the plurality of processing components share a common power supply voltage and clock generator frequency;monitoring at least one thermal parameter of the portable computing device at various locations of the portable computing device external to the plurality of processing components, wherein the at least one thermal parameter of the portable computing device includes a skin temperature of the portable computing device;generating an alarm in response to the monitored thermal parameter of the portable computing device exceeding a predetermined threshold;in response to the alarm, sampling the monitored temperature readings uniquely associated with each of the processing components;based on the sampled temperature readings, querying performance data for each processing component, wherein the performance data represents the relationship between power consumption and workload processing capability for a given individual processing component when operating at a given temperature;comparing the performance data for each processing component to identify a least energy efficient processing component in the multi-processor SoC; andreallocating a first workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the first workload operates to reduce the power consumption by the least energy efficient processing component. 2. The method of claim 1, wherein the least energy efficient processing component is the processing component consuming the most amount of power per workload processed. 3. The method of claim 1, further comprising: determining that the alarm has not cleared;resampling the monitored temperature readings uniquely associated with each of the processing components;based on the resampled temperature readings, re-querying performance data for each processing component;comparing the re-queried performance data for each processing component to identify a new least energy efficient processing component; andreallocating a second workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the second workload operates to reduce the power consumption by the new least energy efficient processing component. 4. The method of claim 1, further comprising: determining that the alarm has cleared; andauthorizing queued workloads to be scheduled to the least energy efficient processing component. 5. The method of claim 1, further comprising: transitioning a power mode of the least energy efficient processor from an active mode to an idle mode. 6. The method of claim 5, further comprising: determining that the alarm has cleared; andauthorizing a return to an active power mode for the least energy efficient processor. 7. The method of claim 1, wherein the thermal parameter is associated with one of a package on package memory temperature, a junction temperature and a battery capacity. 8. The method of claim 1, wherein the portable computing device is in the form of a wireless telephone. 9. A computer system for managing thermal energy generation in a portable computing device having a synchronous multi-processor system on a chip (“SoC”), the system comprising: a hardware monitor module for: monitoring temperature readings uniquely associated with each of a plurality of individual processing components in the multi-processor SoC, wherein the plurality of individual processing components share a common power supply voltage and clock generator frequency;monitoring at least one thermal parameter of the portable computing device at various locations of the portable computing device external to the plurality of processing components, wherein the at least one thermal parameter of the portable computing device includes a skin temperature of the portable computing device;generating an alarm in response to the monitored thermal parameter of the portable computing device exceeding a predetermined threshold; andin response to the alarm, sampling the monitored temperature readings uniquely associated with each of the processing components;an hardware efficiency manager (“EM”) module for: based on the sampled temperature readings, querying performance data for each processing component, wherein the performance data represents the relationship between power consumption and workload processing capability for a given individual processing component when operating at a given temperature;comparing the performance data for each processing component to identify a least energy efficient processing component in the multi-processor SoC; andreallocating a first workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the first workload operates to reduce the power consumption by the least energy efficient processing component. 10. The computer system of claim 9, wherein the least energy efficient processing component is the processing component consuming the most amount of power per workload processed. 11. The computer system of claim 9, wherein: the hardware monitor module is further for: determining that the alarm has not cleared; andresampling the monitored temperature readings uniquely associated with each of the processing components;the hardware EM module is further for: based on the resampled temperature readings, re-querying performance data for each processing component; andcomparing the re-queried performance data for each processing component to identify a new least energy efficient processing component; andreallocating a second workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the second workload operates to reduce the power consumption by the new least energy efficient processing component. 12. The computer system of claim 9, wherein: the hardware monitor module is further for: determining that the alarm has cleared; andthe hardware EM module is further for: authorizing queued workloads to be scheduled to the least energy efficient processing component. 13. The computer system of claim 9, wherein: the hardware monitor module is further for: transitioning a power mode of the least energy efficient processor from an active mode to an idle mode. 14. The computer system of claim 13, wherein: the hardware monitor module is further for: determining that the alarm has cleared; andthe EM module is further for: authorizing a return to an active power mode for the least energy efficient processor. 15. The computer system of claim 9, wherein the thermal parameter is associated with one of a package on package memory temperature, a junction temperature and a battery capacity. 16. A computer system for managing thermal energy generation in a portable computing device having a synchronous multi-processor system on a chip (“SoC”), the system comprising: hardware means for monitoring temperature readings uniquely associated with each of a plurality of individual processing components in the multi-processor SoC, wherein the plurality of individual processing components share a common power supply voltage and clock generator frequency;hardware means for monitoring at least one thermal parameter of the portable computing device at various locations of the portable computing device external to the plurality of processing components, wherein the at least one thermal parameter of the portable computing device includes a skin temperature of the portable computing device;hardware means for generating an alarm in response to the monitored thermal parameter of the portable computing device exceeding a predetermined threshold;hardware means for sampling the monitored temperature readings uniquely associated with each of the processing components in response to the alarm;based on the sampled temperature readings, hardware means for querying performance data for each processing component, wherein the performance data represents the relationship between power consumption and workload processing capability for a given individual processing component when operating at a given temperature;hardware means for comparing the performance data for each processing component to identify a least energy efficient processing component in the multi-processor SoC; andhardware means for reallocating a first workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the first workload operates to reduce the power consumption by the least energy efficient processing component. 17. The computer system of claim 16, wherein the least energy efficient processing component is the processing component consuming the most amount of power per workload processed. 18. The computer system of claim 16, further comprising: hardware means for determining that the alarm has not cleared;hardware means for resampling the monitored temperature readings uniquely associated with each of the processing components;hardware means for re-querying performance data for each processing component based on the resampled temperature readings;hardware means for comparing the re-queried performance data for each processing component to identify a new least energy efficient processing component; andhardware means for reallocating a second workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the second workload operates to reduce the power consumption by the new least energy efficient processing component. 19. The computer system of claim 16, further comprising: hardware means for determining that the alarm has cleared; andhardware means for authorizing queued workloads to be scheduled to the least energy efficient processing component. 20. The computer system of claim 16, further comprising: hardware means for transitioning a power mode of the least energy efficient processor from an active mode to an idle mode. 21. The computer system of claim 20, further comprising: hardware means for determining that the alarm has cleared; andhardware means for authorizing a return to an active power mode for the least energy efficient processor. 22. The computer system of claim 16, wherein the thermal parameter is associated with one of a package on package memory temperature, a junction temperature and a battery capacity. 23. The computer system of claim 16, wherein the portable computing device is in the form of a wireless telephone. 24. A computer program product comprising a non-transitory computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method for managing thermal energy generation in a portable computing device having a synchronous multi-processor system on a chip (“SoC”), said method comprising: monitoring temperature readings uniquely associated with each of a plurality of individual processing components in the multi-processor SoC, wherein the plurality of individual processing components share a common power supply voltage and clock generator frequency;monitoring at least one thermal parameter of the portable computing device at various locations of the portable computing device external to the plurality of processing components, wherein the at least one thermal parameter of the portable computing device includes a skin temperature of the portable computing device;generating an alarm in response to the monitored thermal parameter of the portable computing device exceeding a predetermined threshold;in response to the alarm, sampling the monitored temperature readings uniquely associated with each of the processing components;based on the sampled temperature readings, querying performance data for each processing component, wherein the performance data represents the relationship between power consumption and workload processing capability for a given individual processing component when operating at a given temperature;comparing the performance data for each processing component to identify a least energy efficient processing component in the multi-processor SoC; andreallocating a first workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the first workload operates to reduce the power consumption by the least energy efficient processing component. 25. The computer program product of claim 24, wherein the least energy efficient processing component is the processing component consuming the most amount of power per workload processed. 26. The computer program product of claim 24, further comprising: determining that the alarm has not cleared;resampling the monitored temperature readings uniquely associated with each of the processing components;based on the resampled temperature readings, re-querying performance data for each processing component;comparing the re-queried performance data for each processing component to identify a new least energy efficient processing component; andreallocating a second workload from the least energy efficient processing component to a more energy efficient processing component, wherein reallocating the second workload operates to reduce the power consumption by the new least energy efficient processing component. 27. The computer program product of claim 24, further comprising: determining that the alarm has cleared; andauthorizing queued workloads to be scheduled to the least energy efficient processing component. 28. The computer program product of claim 24, further comprising: transitioning a power mode of the least energy efficient processor from an active mode to an idle mode. 29. The computer program product of claim 28, further comprising: determining that the alarm has cleared; andauthorizing a return to an active power mode for the least energy efficient processor. 30. The computer program product of claim 24, wherein the thermal parameter is associated with one of a package on package memory temperature, a junction temperature and a battery capacity.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (23)
Blanding, William H., Allocating computer resources to workloads using utilization based probability distributions.
Park, Hee-Jun; Thomson, Steven S; Alton, Ronald Frank; Regini, Edoardo; Goverdhan, Satish; Backer, Pieter-Louis Dam, Intelligent multicore control for optimal performance per watt.
Alben, Jonah; Ma, Dennis Kd; Kelleher, Brian, Method and apparatus for power management of graphics processors and subsystems that allow the subsystems to respond to accesses when subsystems are idle.
Berry, Jr., Robert Walter; Johns, Charles Ray; Kuruts, Christopher J., Method and apparatus for power throttling a processor in an information handling system.
Chris S. Browning ; Shekhar Y. Borkar ; Gregory E. Dermer, Method and apparatus for power throttling in a microprocessor using a closed loop feedback system.
Bai, Yu; Vaidya, Priya; Sakarda, Premanand, Method and system for selecting an operating frequency for a chip to provide a desired overall power dissipation value for the chip.
Cohen,Aviad; Finkelstein,Lev; Mendelson,Avi; Ronen,Ronny; Rudoy,Dmitry, Methods and apparatus for optimal voltage and frequency control of thermally limited systems.
Baker, Brian A.; Bandholz, Justin P.; Cox, Jr., William H.; Kochar, Sumeet; Zapata, Ivan R., Operating processor below maximum turbo mode frequency by sending higher than actual current amount signal to monitor.
Capps, Jr., Louis Bennie; Dyckman, Warren D.; Shapiro, Michael Jay, Structure for an apparatus for monitoring and controlling heat generation in a multi-core processor.
Sur, Sumit; Rychlik, Bohuslav; Thomson, Steven S.; Iranli, Ali; Salsbery, Brian J., System and method for dynamically controlling a plurality of cores in a multicore central processing unit based on temperature.
Anderson, Jon J.; Sur, Sumit; Niemann, Jeffrey A.; Artmeier, James M., System and method for managing thermal energy generation in a heterogeneous multi-core processor.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.