发明名称 Energy efficiency aware thermal management in a multi-processor system on a chip
摘要 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.
申请公布号 US9582012(B2) 申请公布日期 2017.02.28
申请号 US201414280629 申请日期 2014.05.18
申请人 QUALCOMM INCORPORATED 发明人 Park Hee Jun;Kang Young Hoon;Alton Ronald Frank;Medrano Christoper Lee;Anderson Jon James
分类号 G06F1/32;G05D23/19;G05B15/02;G06F1/20;G06F1/26 主分类号 G06F1/32
代理机构 Smith Tempel 代理人 Smith Tempel
主权项 1. A method for managing thermal energy generation in a portable computing device having an asynchronous 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 each processing component is associated with a dedicated power supply voltage and clock generator frequency; monitoring at least one thermal parameter 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; and adjusting the dedicated power supply voltage and clock generator frequency of the least energy efficient processing component, wherein adjusting the dedicated power supply voltage and clock generator frequency operates to reduce the power consumption by the least energy efficient processing component, wherein the method further comprises: 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; and adjusting the dedicated power supply voltage and clock generator frequency of the new least energy efficient processing component, wherein adjusting the dedicated power supply voltage and clock generator frequency operates to reduce the power consumption by the new least energy efficient processing component.
地址 San Diego CA US