[go: up one dir, main page]

TWI770830B - Power budget allocations - Google Patents

Power budget allocations Download PDF

Info

Publication number
TWI770830B
TWI770830B TW110106125A TW110106125A TWI770830B TW I770830 B TWI770830 B TW I770830B TW 110106125 A TW110106125 A TW 110106125A TW 110106125 A TW110106125 A TW 110106125A TW I770830 B TWI770830 B TW I770830B
Authority
TW
Taiwan
Prior art keywords
workload
power budget
power
component
budget
Prior art date
Application number
TW110106125A
Other languages
Chinese (zh)
Other versions
TW202205056A (en
Inventor
方洋 戴
思遠 張
喬恩 G 洛伊德
琪竣 陳
阿斯賈德 夏米恩
Original Assignee
美商惠普發展公司有限責任合夥企業
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 美商惠普發展公司有限責任合夥企業 filed Critical 美商惠普發展公司有限責任合夥企業
Publication of TW202205056A publication Critical patent/TW202205056A/en
Application granted granted Critical
Publication of TWI770830B publication Critical patent/TWI770830B/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26Power supply means, e.g. regulation thereof
    • G06F1/32Means for saving power
    • G06F1/3203Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3206Monitoring of events, devices or parameters that trigger a change in power modality
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26Power supply means, e.g. regulation thereof
    • G06F1/32Means for saving power
    • G06F1/3203Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3234Power saving characterised by the action undertaken
    • G06F1/3243Power saving in microcontroller unit
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

An example computing device includes: a first component having a first power budget; a second component having a second power budget; and a power allocation controller to: obtain (i) a first power consumption indicator for the first component, and (ii) a second power consumption indicator for the second component; generate (i) a first workload estimate based on the first power budget and the first power consumption indicator, and (ii) a second workload estimate based on the second power budget and the second power consumption indicator; and provide, based on a comparison of the first and second workload estimates, (i) a first adjusted power budget to the first component, and (ii) a second adjusted power budget to the second component.

Description

功率預算分配技術Power Budget Allocation Technology

本發明係有關於功率預算分配技術。The present invention relates to power budget allocation techniques.

發明背景Background of the Invention

計算裝置可例如基於功率供應及/或該裝置之冷卻能力而將功率預算分配至其某些組件。此類功率預算之分配可在計算上密集,及/或可由於對一或多個組件之過度限制性預算分配而導致裝置效能下降。A computing device may allocate a power budget to certain of its components, for example, based on power supply and/or cooling capabilities of the device. Allocation of such power budgets can be computationally intensive and/or can result in device performance degradation due to overly restrictive budget allocations to one or more components.

依據本發明之一實施例,係特地提出一種計算裝置,其包含:一第一組件,其具有一第一功率預算;一第二組件,其具有一第二功率預算;以及一功率分配控制器,其用以:獲得(i)該第一組件之一第一功率消耗指示符,及(ii)該第二組件之一第二功率消耗指示符;基於該第一功率預算及該第一功率消耗指示符產生(i)一第一工作負載估計,且基於該第二功率預算及該第二功率消耗指示符產生(ii)一第二工作負載估計;以及基於該第一工作負載估計與該第二工作負載估計之一比較,將(i)一第一經調整功率預算提供至該第一組件且將(ii)一第二經調整功率預算提供至該第二組件。According to an embodiment of the present invention, a computing device is specially proposed, which includes: a first component having a first power budget; a second component having a second power budget; and a power distribution controller , which is used to: obtain (i) a first power consumption indicator for the first component, and (ii) a second power consumption indicator for the second component; based on the first power budget and the first power generating (i) a first workload estimate based on the consumption indicator, and generating (ii) a second workload estimate based on the second power budget and the second power consumption indicator; and generating (ii) a second workload estimate based on the first workload estimate and the A comparison of the second workload estimate provides (i) a first adjusted power budget to the first component and (ii) a second adjusted power budget to the second component.

較佳實施例之詳細說明DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

圖1展示實例計算裝置100。計算裝置100可為桌上型電腦、伺服器、智慧型手機、筆記本電腦、平板電腦或類似裝置。計算裝置100包括第一目標組件104-1及第二目標組件104-2 (其可統稱為目標組件104,且通常統稱為一目標組件104)。目標組件104-1及104-2亦可簡稱為組件104-1及104-2。在一些實例中,第一目標組件104-1包括中央處理單元(CPU),且第二目標組件104-2包括圖形處理單元(GPU)。計算裝置100可包括其他組件,諸如儲存裝置(例如,硬碟機、依電性記憶體以及類似者)、輸入裝置(例如,鍵盤)及輸出裝置(例如,顯示器)。FIG. 1 shows an example computing device 100 . Computing device 100 may be a desktop computer, server, smartphone, laptop, tablet, or similar device. Computing device 100 includes a first target component 104-1 and a second target component 104-2 (which may be collectively referred to as target components 104, and are often collectively referred to as a target component 104). The target components 104-1 and 104-2 may also be simply referred to as components 104-1 and 104-2. In some examples, the first target component 104-1 includes a central processing unit (CPU) and the second target component 104-2 includes a graphics processing unit (GPU). Computing device 100 may include other components such as storage devices (eg, hard drives, electrical memory, and the like), input devices (eg, keyboards), and output devices (eg, displays).

利用例如來自電源108之功率供應計算裝置100之組件,包括目標組件104。電源108可為與計算裝置100整合之電池,諸如膝上型電池。在其他實例中,電源108可在計算裝置100外部,諸如外部電池、電氣實用連接件或類似者。Components of computing device 100 , including target component 104 , are supplied with power, such as from power source 108 . Power source 108 may be a battery integrated with computing device 100, such as a laptop battery. In other examples, the power source 108 may be external to the computing device 100, such as an external battery, electrical utility connections, or the like.

計算裝置100亦包括用以將功率預算分配至目標組件104之功率分配控制器112 (其亦可稱為控制器112)。控制器112可實施為例如計算裝置100之主板上的嵌式控制器(EC)。Computing device 100 also includes a power distribution controller 112 (which may also be referred to as controller 112 ) for allocating the power budget to target components 104 . Controller 112 may be implemented, for example, as an embedded controller (EC) on a motherboard of computing device 100 .

目標組件104可用於藉由功率分配控制器112基於例如由目標組件104表示之計算裝置100之總功率消耗之比例進行的功率預算分配。換言之,諸如CPU及GPU之組件可表示計算裝置100之總功率消耗的相當大部分。舉例而言,在膝上型電腦中,CPU及GPU一起可表示超過電腦之總功率消耗的一半。經由目標組件104之功率消耗施加控制可因此使得功率分配控制器112能夠管理裝置100之總功率消耗,以滿足可用於計算裝置100的冷卻及/或功率遞送能力。The target component 104 may be used for power budget allocation by the power allocation controller 112 based on, for example, a proportion of the total power consumption of the computing device 100 represented by the target component 104 . In other words, components such as the CPU and GPU may represent a significant portion of the total power consumption of computing device 100 . For example, in a laptop, the CPU and GPU together may represent more than half of the total power consumption of the computer. Applying control over the power consumption of the target components 104 may thus enable the power distribution controller 112 to manage the overall power consumption of the device 100 to meet the cooling and/or power delivery capabilities available for the computing device 100 .

計算裝置100進一步包括耦接至功率分配控制器112之非暫時性電腦可讀媒體,諸如儲存裝置116,其亦可稱為記憶體。在其他實例中,儲存裝置116亦可與控制器112整合。儲存裝置116含有指令集118,其可由控制器112執行以實施下文更詳細地論述的功率預算分配功能。Computing device 100 further includes a non-transitory computer-readable medium, such as storage device 116 , which may also be referred to as memory, coupled to power distribution controller 112 . In other examples, the storage device 116 may also be integrated with the controller 112 . The storage device 116 contains a set of instructions 118 executable by the controller 112 to implement the power budget allocation functions discussed in more detail below.

控制器112連接至自電源108至目標組件104之功率供應導管,在圖1中以實線展示。控制器112亦連接至儲存裝置116及目標組件104以用於資料通訊,如以虛線展示。如下文將詳細地論述,控制器112獲得目標組件104-1之經量測功率消耗及目標組件104-2之經量測功率消耗。上述經量測功率消耗亦可稱為功率消耗指示符。舉例而言,組件104-1之功率消耗指示符可包括組件104-1之在諸如一秒之預定義時間段內的以瓦特(Watt)為單位的平均功率消耗。亦預期短於及長於一秒之其他時間段。亦即,控制器112可在時間段內獲得多個量測值,且可根據該多個量測值產生指示符(例如,平均功率消耗)。在其他實例中,組件104-1之功率消耗指示符可包括在時間段內由組件104-1消耗之例如以焦耳(Joule)為單位之能量的量。The controller 112 is connected to a power supply conduit from the power source 108 to the target device 104 , shown in solid lines in FIG. 1 . Controller 112 is also connected to storage device 116 and target component 104 for data communication, as shown in dashed lines. As will be discussed in detail below, the controller 112 obtains the measured power consumption of the target component 104-1 and the measured power consumption of the target component 104-2. The above-mentioned measured power consumption may also be referred to as a power consumption indicator. For example, the power consumption indicator for component 104-1 may include the average power consumption in Watts of component 104-1 over a predefined time period, such as one second. Other time periods shorter and longer than one second are also contemplated. That is, the controller 112 may obtain a plurality of measurement values over a period of time, and may generate an indicator (eg, average power consumption) according to the plurality of measurement values. In other examples, the power consumption indicator for component 104-1 may include the amount of energy, eg, in Joules, consumed by component 104-1 over a period of time.

在獲得上文所提及之功率消耗指示符後,控制器112可產生第一目標組件104-1之第一工作負載估計及第二目標組件104-2之第二工作負載估計。控制器112可基於對應於彼組件104之功率消耗指示符且基於分配至彼組件104之當前功率預算而產生組件104之工作負載估計。因此,舉例而言,控制器112可將40 W之功率預算分配至目標組件104-1 (例如,CPU),且獲得指示在前一秒時間段內25 W之平均功率消耗的功率消耗指示符。可判定組件104-1之工作負載估計為功率消耗指示符與所分配功率預算之比率。因此,在上述實例中,工作負載估計因此為0.625。更一般而言,工作負載估計可為零與一之間的分數,但亦可使用其他格式來表示工作負載估計。After obtaining the power consumption indicators mentioned above, the controller 112 may generate a first workload estimate for the first target component 104-1 and a second workload estimate for the second target component 104-2. The controller 112 may generate a workload estimate for the component 104 based on the power consumption indicator corresponding to that component 104 and based on the current power budget allocated to that component 104 . Thus, for example, the controller 112 may allocate a power budget of 40 W to the target component 104-1 (eg, CPU) and obtain a power consumption indicator indicating an average power consumption of 25 W over the previous second period . The workload estimate of the decidable component 104-1 is the ratio of the power consumption indicator to the allocated power budget. Therefore, in the above example, the workload estimate is thus 0.625. More generally, workload estimates can be fractions between zero and one, although other formats can also be used to represent workload estimates.

換言之,由控制器112產生之工作負載估計指示給定目標組件104之功率預算的多少部分在相關時間段(例如,在產生工作負載估計之前的一秒時間段)內由彼目標組件104使用。In other words, the workload estimate generated by controller 112 indicates how much of a given target component 104's power budget was used by that target component 104 during the relevant time period (eg, the one-second period before the workload estimate was generated).

控制器112可接著將對應於第一目標組件104-1之第一工作負載估計與對應於第二目標組件104-2之第二工作負載估計進行比較。基於該比較,控制器112可將經調整功率預算提供至第一目標組件104-1,且將經調整功率預算提供至第二目標組件104-2。分配至目標組件104的經調整功率預算可包括先前所分配功率預算的增大或減小。亦即,一個目標組件104 (例如,第一目標組件104-1)之功率預算之一部分可重新分配至另一目標組件104 (例如,第二目標組件104-2)。The controller 112 may then compare the first workload estimate corresponding to the first target component 104-1 with the second workload estimate corresponding to the second target component 104-2. Based on the comparison, the controller 112 may provide the adjusted power budget to the first target component 104-1 and the adjusted power budget to the second target component 104-2. The adjusted power budget allocated to the target component 104 may include an increase or decrease in the previously allocated power budget. That is, a portion of the power budget of one target device 104 (eg, first target device 104-1) may be reallocated to another target device 104 (eg, second target device 104-2).

自一個目標組件104至另一目標組件104重新分配之功率預算之部分可定義為例如系統功率預算之一部分,其表示目標組件104之經組合功率預算。舉例而言,對於100 W之系統功率預算,至第一目標組件104-1之預設功率預算分配可為40 W,且至第二目標組件104-2之預設功率預算分配可為60 W。回應於對組件104之工作負載估計之上文所提及的比較,控制器112可將系統功率預算之5% (在此實例中,5 W)自第一目標組件104-1重新分配至第二目標組件104-2。第一經調整功率預算因此為35 W,且第二經調整功率預算為65 W。The portion of the power budget that is reallocated from one target device 104 to another target device 104 may be defined, for example, as a portion of the system power budget that represents the combined power budget of the target devices 104 . For example, for a system power budget of 100 W, the default power budget allocation to the first target device 104-1 may be 40 W, and the default power budget allocation to the second target device 104-2 may be 60 W . In response to the aforementioned comparison of the workload estimates for component 104, controller 112 may reallocate 5% of the system power budget (in this example, 5 W) from the first target component 104-1 to the second target component 104-1. Two target components 104-2. The first adjusted power budget is thus 35 W and the second adjusted power budget is 65 W.

在其他實例中,目標組件104之間重新分配的功率預算之部分可以絕對瓦數而非作為系統功率預算之分數來指定。重新分配的功率預算之部分亦可基於第一工作負載估計與第二工作負載估計之間的比較而在大小方面發生變化。亦即,可回應於工作負載估計之間的較大差而重新分配功率預算之較大部分。相反地,可回應於工作負載估計之間的較小差而重新分配功率預算之較小部分(包括少至零重新分配)。下文將論述重新分配判定的某些實例。In other examples, the portion of the power budget that is reallocated among the target components 104 may be specified in absolute wattages rather than as a fraction of the system power budget. Portions of the reallocated power budget may also vary in size based on the comparison between the first workload estimate and the second workload estimate. That is, a larger portion of the power budget may be reallocated in response to larger differences between workload estimates. Conversely, smaller portions of the power budget (including as little as zero reallocation) may be reallocated in response to smaller differences between workload estimates. Some examples of reallocation decisions are discussed below.

控制器112可重複上述分配過程,例如每秒一次或根據另一適合的時間段。控制器112可因此根據目標組件104之當前工作負載而動態地調整目標組件104之功率預算分配。此類動態調整可增大功率預算分配至目標組件104之效率,例如降低一個目標組件104調節計算活動以避免超出所分配功率預算之情況的頻率,從而潛在地降低效能,而另一目標組件104利用至彼目標組件104之功率預算分配的一小部分。藉由功率分配控制器112實施之動態調整可因此能夠提昇計算裝置100之效能。The controller 112 may repeat the above-described dispensing process, eg, once per second or according to another suitable time period. The controller 112 can thus dynamically adjust the power budget allocation of the target device 104 according to the current workload of the target device 104 . Such dynamic adjustments can increase the efficiency of power budget allocation to target components 104 , eg, reducing the frequency with which one target component 104 adjusts computational activity to avoid situations where the allocated power budget is exceeded, potentially reducing performance, while another target component 104 A fraction of the power budget allocation to that target device 104 is utilized. The dynamic adjustment implemented by the power distribution controller 112 may thus improve the performance of the computing device 100 .

圖2展示分配計算組件之功率預算之實例方法200。方法200可由指令集實施,該指令集儲存於非暫時性電腦可讀媒體中(例如,儲存於儲存裝置116中的指令118,如圖1中所展示)且由諸如功率分配控制器112的控制器執行。下文結合由計算裝置100進行的方法200之實例執行描述方法200。2 shows an example method 200 of allocating a power budget for computing components. Method 200 may be implemented by a set of instructions stored in a non-transitory computer-readable medium (eg, instructions 118 stored in storage device 116 , as shown in FIG. 1 ) and controlled by, for example, power distribution controller 112 device executes. The method 200 is described below in conjunction with an example performance of the method 200 performed by the computing device 100 .

在區塊205處,控制器112設定目標組件104之初始功率預算,其亦可稱為預設功率預算。初始功率預算可設定為系統功率預算之預定部分。系統功率預算自身之判定可基於諸如剩餘電池容量、環境溫度以及類似者之因素而變化。在本實例中,假設系統功率預算為100 W,且將忽略系統功率預算之此可變性。進一步假設例如CPU之第一目標組件104-1接收對應於40%之系統功率預算或40 W的初始功率預算。例如GPU之第二目標組件104-2因此接收對應於60%之系統功率預算或60 W的初始功率預算。下文在表1中說明第一目標組件104-1之第一初始功率預算及第二目標組件104-2之第二初始功率預算。 表1:初始功率預算(W) 系統 CPU GPU 100 40 60 At block 205, the controller 112 sets an initial power budget for the target device 104, which may also be referred to as a preset power budget. The initial power budget can be set as a predetermined portion of the system power budget. The determination of the system power budget itself may vary based on factors such as remaining battery capacity, ambient temperature, and the like. In this example, the system power budget is assumed to be 100 W, and this variability in system power budget will be ignored. Assume further that the first target component 104-1, eg, a CPU, receives a system power budget corresponding to 40% or an initial power budget of 40 W. The second target component 104-2, eg, a GPU, thus receives a system power budget corresponding to 60% or an initial power budget of 60 W. The first initial power budget for the first target device 104-1 and the second initial power budget for the second target device 104-2 are described in Table 1 below. Table 1: Initial Power Budget (W) system CPU GPU 100 40 60

在區塊210處,控制器112可獲得第一目標組件104-1之第一功率消耗指示符及第二目標組件104-2之第二功率消耗指示符。舉例而言,控制器112可在時間段(例如,一秒)之進程內獲得目標組件104-1及104-2之當前功率消耗的多個量測值,且基於該多個量測值產生功率消耗指示符。出於例示性目的,假設在區塊210處產生之第一功率消耗指示符(針對目標組件104-1)指示在區塊210之執行之前的一秒時段內之38 W的平均功率消耗,且第二功率消耗指示符(針對目標組件104-2)指示在同一一秒時間段內之15 W的平均功率消耗。At block 210, the controller 112 may obtain a first power consumption indicator for the first target component 104-1 and a second power consumption indicator for the second target component 104-2. For example, the controller 112 may obtain multiple measurements of the current power consumption of the target components 104-1 and 104-2 over the course of a time period (eg, one second), and generate based on the multiple measurements Power consumption indicator. For illustrative purposes, assume that the first power consumption indicator (for target component 104-1) generated at block 210 indicates an average power consumption of 38 W over the one-second period prior to execution of block 210, and A second power consumption indicator (for target component 104-2) indicates an average power consumption of 15 W over the same one second period.

在區塊215處,控制器112可產生第一目標組件104-1之第一工作負載估計及第二目標組件104-2之第二工作負載估計。如前面所指出,工作負載估計可產生為功率消耗指示符與當前功率預算之比率。因此,在本實例中,第一目標組件104-1之第一工作負載指示符為35 W與40 W之比率,或0.95。第二目標組件104-2之第二工作負載指示符為15 W與60 W之比率,或0.25。表2匯總方法200之本實例執行中的初始功率預算、功率消耗及工作負載估計。 表2:功率消耗指示符及工作負載估計   CPU GPU 初始預算 ( W ) 40 60 功率消耗 ( W ) 38 15 工作負載估計 0.95 0.25 At block 215, the controller 112 may generate a first workload estimate for the first target component 104-1 and a second workload estimate for the second target component 104-2. As noted earlier, the workload estimate may be generated as a ratio of a power consumption indicator to the current power budget. Thus, in this example, the first workload indicator for the first target component 104-1 is the ratio of 35 W to 40 W, or 0.95. The second workload indicator for the second target component 104-2 is the ratio of 15 W to 60 W, or 0.25. Table 2 summarizes the initial power budget, power consumption, and workload estimates in this example implementation of method 200. Table 2: Power consumption indicators and workload estimates CPU GPU Initial Budget ( W ) 40 60 Power consumption ( W ) 38 15 Workload estimation 0.95 0.25

在區塊220處,控制器112可比較工作負載估計,且基於該比較將經調整功率預算提供至第一目標組件104-1及第二目標組件104-2。At block 220, the controller 112 may compare the workload estimates and provide an adjusted power budget to the first target component 104-1 and the second target component 104-2 based on the comparison.

在一些實例中,控制器112可例如藉由判定第一工作負載估計與第二工作負載估計之間的差來直接比較工作負載估計。在本實例中,第一工作負載估計與第二工作負載估計之間的差為0.7。在判定上述差後,控制器112可選擇第一初始功率預算及第二初始功率預算中之一者之一部分來進行重新分配。In some examples, the controller 112 may directly compare the workload estimates, eg, by determining the difference between the first workload estimate and the second workload estimate. In this example, the difference between the first workload estimate and the second workload estimate is 0.7. After determining the difference, the controller 112 may select a portion of one of the first initial power budget and the second initial power budget for reallocation.

待重新分配的部分可與工作負載估計之間的差成比例,直至系統預算之預定部分。舉例而言,控制器112可在系統功率預算之0%與15%之間選擇用於重新分配的一部分,根據該差而按比例調整。在本實例中,第一工作負載估計與第二工作負載估計之間的差為0.7,且控制器112可因此選擇10.5% (亦即,15%之0.7)之一部分以進行重新分配。表3展示此重新分配之後的第一目標組件104-1及第二目標組件104-2之經調整功率預算。特定言之,已向第一目標組件104-1分配額外10.5 W (亦即,系統預算之10.5%),且已向第二目標組件分配小於10.5 W。 表3:功率消耗指示符、工作負載估計、經調整功率預算   CPU GPU 初始預算 ( W ) 40 60 功率消耗 ( W ) 38 15 工作負載估計 0.95 0.25 經調整功率預算 ( W ) 50.5 49.5 The portion to be reallocated may be proportional to the difference between workload estimates, up to a predetermined portion of the system budget. For example, the controller 112 may select a portion for reallocation between 0% and 15% of the system power budget, proportionally adjusted according to the difference. In this example, the difference between the first workload estimate and the second workload estimate is 0.7, and the controller 112 may therefore select a fraction of 10.5% (ie, 0.7 of 15%) for reallocation. Table 3 shows the adjusted power budgets of the first target device 104-1 and the second target device 104-2 after this reallocation. In particular, an additional 10.5 W (ie, 10.5% of the system budget) has been allocated to the first target component 104-1, and less than 10.5 W has been allocated to the second target component. Table 3: Power Consumption Indicators, Workload Estimation, Adjusted Power Budget CPU GPU Initial Budget ( W ) 40 60 Power consumption ( W ) 38 15 Workload estimation 0.95 0.25 Adjusted Power Budget ( W ) 50.5 49.5

如對於熟習此項技術者現將顯而易見,在區塊220處提供經調整功率預算之後,第一目標組件104-1接收增加之功率預算以反映第一目標組件104-1先前比第二目標組件104-2利用可用功率之更大部分的事實。藉由使更大功率量可用於第一目標組件104-1,可提昇或維持第一目標組件104-1之效能。同時,分配更少功率至第二目標組件104-2可對目標組件104-2之效能產生極小影響。As will now be apparent to those skilled in the art, after providing the adjusted power budget at block 220, the first target component 104-1 receives an increased power budget to reflect that the first target component 104-1 was previously larger than the second target component The fact that 104-2 utilizes a larger portion of the available power. By making a greater amount of power available to the first target device 104-1, the performance of the first target device 104-1 can be enhanced or maintained. At the same time, allocating less power to the second target device 104-2 may have little impact on the performance of the target device 104-2.

在執行區塊220後,控制器112可接著返回至區塊210以評估下一時間段(例如,下一一秒時段)之功率消耗。控制器112可因此回應於改變目標組件104之工作負載而連續地更新目標組件104之所分配功率預算,如由功率消耗及當前預算分配所判定。區塊210、215及220之每一執行可導致選擇系統功率預算之一部分以自第一初始功率預算及第二初始功率預算中之一者重新分配至另一者。After performing block 220, the controller 112 may then return to block 210 to evaluate the power consumption for the next period of time (eg, the next one-second period). The controller 112 may thus continuously update the allocated power budget of the target components 104 in response to changing the workload of the target components 104, as determined by the power consumption and the current budget allocation. Each execution of blocks 210, 215, and 220 may result in selection of a portion of the system power budget for reallocation from one of the first and second initial power budgets to the other.

表4說明在上文所論述之第一實例執行之後的區塊21、215及220之另一實例執行的結果。特定言之,假設區塊210處針對第一目標組件104-1量測的功率消耗為40 W,且針對第二目標組件104-1量測的功率消耗為27 W。亦即,第一目標組件104-1之功率消耗增加3 W,而第二目標組件104-2之功率消耗已幾乎加倍。 表4:功率消耗指示符、工作負載估計、經調整功率預算   CPU GPU 初始預算 ( W ) 40 60 先前預算 ( W ) 50.5 49.5 功率消耗 ( W ) 40 27 工作負載估計 0.79 0.54 經調整功率預算 ( W ) 43.75 56.25 Table 4 illustrates the results of another example execution of blocks 21, 215, and 220 following execution of the first example discussed above. Specifically, it is assumed that the measured power consumption at block 210 for the first target device 104-1 is 40 W, and the measured power consumption for the second target device 104-1 is 27 W. That is, the power consumption of the first target device 104-1 has increased by 3 W, while the power consumption of the second target device 104-2 has almost doubled. Table 4: Power Consumption Indicators, Workload Estimation, Adjusted Power Budget CPU GPU Initial Budget ( W ) 40 60 previous budget ( W ) 50.5 49.5 Power consumption ( W ) 40 27 Workload estimation 0.79 0.54 Adjusted Power Budget ( W ) 43.75 56.25

在區塊215處產生之工作負載估計對於第一目標組件104-1為0.79且對於第二目標組件104-2為0.54。工作負載估計之間的差為0.25,且待重新分配之系統預算之部分因此為3.75% (15%之0.25),在此實例中等同於3.75 W。因此,經調整功率預算對於第一目標組件104-1為43.75 W且對於第二目標組件104-2為56.25 W。The workload estimate generated at block 215 is 0.79 for the first target component 104-1 and 0.54 for the second target component 104-2. The difference between the workload estimates is 0.25, and the portion of the system budget to be reallocated is therefore 3.75% (0.25 of 15%), which in this example equates to 3.75 W. Thus, the adjusted power budget is 43.75 W for the first target component 104-1 and 56.25 W for the second target component 104-2.

亦預期其他機制用於在區塊220處執行比較及所得預算分配調整。舉例而言,替代直接比較第一工作負載估計與第二工作負載估計,控制器112可將第一工作負載估計與臨限值進行比較以產生第一工作負載狀態指示符,且可將第二工作負載估計與臨限值進行比較以產生第二工作負載狀態指示符。在一些實例中,控制器112可採用超過一個臨限值。換言之,可採用臨限值來將工作負載估計離散為一組工作負載狀態指示符中之一者,例如指示目標組件104之工作負載為低、中抑或高。Other mechanisms are also contemplated for performing the comparison and resulting budget allocation adjustment at block 220 . For example, instead of directly comparing the first workload estimate to the second workload estimate, the controller 112 may compare the first workload estimate to a threshold value to generate a first workload status indicator, and may compare the second workload estimate to a threshold value. The workload estimate is compared to a threshold value to generate a second workload status indicator. In some instances, the controller 112 may employ more than a threshold value. In other words, a threshold value may be employed to discretize the workload estimate into one of a set of workload status indicators, eg, indicating whether the workload of the target component 104 is low, medium, or high.

圖3展示分配計算組件之功率預算之另一實例方法300。方法300如上文結合方法200之區塊205、210及215所描述開始。在區塊305處,控制器112將第一目標組件104-1之第一工作負載估計與至少一個臨限值進行比較,且亦將第二目標組件104-2之第二工作負載估計與臨限值進行比較。在本實例中,控制器112可將工作負載估計與上臨限值及下臨限值進行比較。基於與上及下臨限值之比較,控制器112產生第一目標組件104-1之第一工作負載狀態指示符及第二目標組件104-2之第二工作負載狀態指示符。3 shows another example method 300 of allocating a power budget for computing components. Method 300 begins as described above in connection with blocks 205 , 210 and 215 of method 200 . At block 305, the controller 112 compares the first workload estimate for the first target component 104-1 to at least one threshold value, and also compares the second workload estimate for the second target component 104-2 with the limit for comparison. In this example, the controller 112 may compare the workload estimate to an upper threshold value and a lower threshold value. Based on the comparison with the upper and lower thresholds, the controller 112 generates a first workload status indicator for the first target component 104-1 and a second workload status indicator for the second target component 104-2.

上文所提及之上及下臨限值可設定為各種值。在下文論述之方法300之例示性執行中,下臨限值為0.25且上臨限值為0.75。低於下臨限值之工作負載估計產生為「低」之工作負載狀態指示符,而下臨限值與上臨限值之間的工作負載估計產生為「中」之工作負載狀態指示符,且高於上臨限值之工作負載估計產生為「高」之工作負載狀態指示符。亦可採用各種其他形式之工作負載狀態指示符。舉例而言,在其他實例中,可使用數值指示符,諸如對於低為「0」、對於中為「1」以及對於高為「2」。The above-mentioned upper and lower threshold values can be set to various values. In an exemplary implementation of method 300 discussed below, the lower threshold value is 0.25 and the upper threshold value is 0.75. A workload estimate below the lower threshold value yields a workload status indicator of "low", while workload estimates between the lower threshold value and the upper threshold value yield a workload status indicator of "medium", And workload estimates above the upper threshold value generate a workload status indicator of "high". Various other forms of workload status indicators may also be employed. For example, in other examples, numerical indicators may be used, such as "0" for low, "1" for medium, and "2" for high.

表5說明第一及第二目標組件104之實例功率消耗指示符、工作負載估計及工作負載狀態指示符。 表5:功率消耗指示符、工作負載估計、經調整功率預算   CPU GPU 初始預算 ( W ) 40 60 功率消耗 ( W ) 25 50 工作負載估計 0.625 0.83 工作負載狀態指示符 經調整功率預算 ( W ) 33 67 Table 5 illustrates example power consumption indicators, workload estimates, and workload status indicators for the first and second target components 104 . Table 5: Power Consumption Indicators, Workload Estimation, Adjusted Power Budget CPU GPU Initial Budget ( W ) 40 60 Power consumption ( W ) 25 50 Workload estimation 0.625 0.83 Workload Status Indicator middle high Adjusted Power Budget ( W ) 33 67

如表5中所見,第一目標組件104-1之工作負載估計處於上與下臨限值之間,且因此選擇工作負載狀態指示符「中」。第二目標組件104-2之工作負載估計高於上臨限值,且因此選擇工作負載狀態指示符「高」。As seen in Table 5, the workload estimate for the first target component 104-1 is between the upper and lower thresholds, and the workload status indicator "Medium" is therefore selected. The workload estimate for the second target component 104-2 is above the upper threshold, and the workload status indicator "high" is therefore selected.

為產生第一及第二目標組件104之經調整功率預算,控制器112進行至區塊310。在區塊310處,控制器112判定工作負載狀態指示符之間的差是等於抑或低於臨限值。在一些實例中,臨限值可為零,且因此控制器112可在區塊310處判定工作負載狀態指示符是否相等。當區塊310處之判定為肯定的(例如,若二個工作負載狀態指示符為「中」)時,控制器112進行至區塊315,且將目標組件104之經調整功率預算設定為初始功率預算。換言之,當工作負載狀態指示符相等時,不對初始(預設)功率預算作出改變。To generate adjusted power budgets for the first and second target components 104 , the controller 112 proceeds to block 310 . At block 310, the controller 112 determines whether the difference between the workload status indicators is equal to or below a threshold value. In some examples, the threshold value may be zero, and thus the controller 112 may determine at block 310 whether the workload status indicators are equal. When the determination at block 310 is positive (eg, if both workload status indicators are "medium"), the controller 112 proceeds to block 315 and sets the adjusted power budget of the target device 104 to initial power budget. In other words, when the workload status indicators are equal, no changes are made to the initial (preset) power budget.

另一方面,當工作負載狀態指示符之間的差超過臨限值時(例如,當工作負載狀態指示符並不相等時),控制器112進行至區塊320。在區塊320處,控制器112將一個目標組件104之功率預算之一部分重新分配至另一目標組件104。舉例而言,控制器112可藉由根據工作負載狀態指示符之間的差添加或減去系統預算之預定部分來調整目標組件功率預算。舉例而言,當工作負載狀態指示符相差二級(例如,「高」及「低」時),控制器112可將系統預算之15%自一個目標組件104重新分配至另一者。當工作負載狀態指示符相差一級(例如,「高」及「中間」,或「中間」及「低」)時,控制器112可將系統預算之7%自一個目標組件104重新分配至另一者。On the other hand, when the difference between the workload status indicators exceeds a threshold value (eg, when the workload status indicators are not equal), the controller 112 proceeds to block 320 . At block 320 , the controller 112 reallocates a portion of the power budget of one target device 104 to another target device 104 . For example, the controller 112 may adjust the target device power budget by adding or subtracting a predetermined portion of the system budget based on the difference between the workload status indicators. For example, when the workload status indicators differ by two levels (eg, "high" and "low"), the controller 112 may reallocate 15% of the system budget from one target component 104 to the other. When the workload status indicators differ by one level (eg, "high" and "middle", or "middle" and "low"), the controller 112 may reallocate 7% of the system budget from one target component 104 to another By.

因此,在方法300之本實例執行中,控制器112可將7% (或7 W,在系統預算為100 W之情況下)自第一目標組件104-1重新分配至第二目標組件104-2,從而導致33 W及67 W之經調整功率預算,如表5中所展示。在區塊325處,控制器112可例如經由先前提及且在圖1中展示的資料通訊介面將經調整功率預算提供至目標組件104-1及104-2。控制器112可接著返回至區塊210以在下一時間段重複上述過程。Thus, in this example implementation of method 300, controller 112 may reallocate 7% (or 7 W, if the system budget is 100 W) from first target component 104-1 to second target component 104- 2, resulting in adjusted power budgets of 33 W and 67 W, as shown in Table 5. At block 325, the controller 112 may provide the adjusted power budget to the target components 104-1 and 104-2, eg, via the data communication interface previously mentioned and shown in FIG. 1 . The controller 112 may then return to block 210 to repeat the above process for the next time period.

可直接獲得由控制器112獲得的經量測功率消耗,如圖1中所展示。亦即,計算裝置100可包括與功率分配控制器112整合之功率消耗量測裝置。The measured power consumption obtained by the controller 112 can be obtained directly, as shown in FIG. 1 . That is, the computing device 100 may include a power consumption measurement device integrated with the power distribution controller 112 .

在其他實例中,控制器112可自另一組件接收經量測功率消耗。圖4展示另一實例計算裝置400,包括目標組件104-1及104-2、電源108、功率分配控制器112及儲存裝置116。然而,在圖4的實例中,計算裝置400包括與控制器112分別地實施的量測裝置404。量測裝置404可為例如連接至電源108與目標組件104之間的功率導管的能量估計引擎(E3)裝置。量測裝置404亦連接至控制器112以用於資料通訊,且將功率消耗量測值例如連續地提供至控制器112。In other examples, the controller 112 may receive the measured power consumption from another component. FIG. 4 shows another example computing device 400 including target components 104 - 1 and 104 - 2 , power source 108 , power distribution controller 112 , and storage device 116 . However, in the example of FIG. 4 , computing device 400 includes measurement device 404 implemented separately from controller 112 . The measurement device 404 may be, for example, an energy estimation engine (E3) device connected to a power conduit between the power source 108 and the target assembly 104 . The measurement device 404 is also connected to the controller 112 for data communication and provides power consumption measurements to the controller 112, eg, continuously.

應認識到,上文提供之各種實例之特徵及態樣可組合至亦屬於本揭露內容之範疇之其他實例中。另外,出於例示性目的,圖式並不按比例且可放大大小及形狀。It should be appreciated that the features and aspects of the various examples provided above may be combined into other examples that are also within the scope of the present disclosure. Additionally, the drawings are not to scale and may be exaggerated in size and shape for illustrative purposes.

100,400:計算裝置 104,104-1,104-2:目標組件 108:電源 112:功率分配控制器 116:儲存裝置 118:指令集 200,300:方法 205,210,215,220,305,310,315,320,325:區塊 404:量測裝置100,400: Computing device 104, 104-1, 104-2: Target components 108: Power 112: Power distribution controller 116: Storage Device 118: Instruction set 200,300: Method 205, 210, 215, 220, 305, 310, 315, 320, 325: Blocks 404: Measuring device

圖1為具有用以將功率預算分配至計算組件之功率分配控制器之實例計算裝置的區塊圖。1 is a block diagram of an example computing device with a power allocation controller to allocate a power budget to computing components.

圖2為將功率預算分配至計算裝置中之計算組件之實例方法的流程圖。2 is a flowchart of an example method of allocating a power budget to computing components in a computing device.

圖3為將功率預算分配至計算裝置中之計算組件之另一實例方法的流程圖。3 is a flowchart of another example method of allocating a power budget to computing components in a computing device.

圖4為具有用以將功率預算分配至計算組件之功率分配控制器之另一實例計算裝置的區塊圖。4 is a block diagram of another example computing device with a power allocation controller to allocate power budgets to computing components.

100:計算裝置100: Computing Devices

104-1,104-2:目標組件104-1, 104-2: Target Components

108:電源108: Power

112:功率分配控制器112: Power distribution controller

116:儲存裝置116: Storage Device

118:指令集118: Instruction set

Claims (15)

一種計算裝置,其包含:一第一組件,其具有一第一功率預算;一第二組件,其具有一第二功率預算;以及一功率分配控制器,其用以:獲得(i)該第一組件之一第一功率消耗指示符,及(ii)該第二組件之一第二功率消耗指示符;基於該第一功率預算及該第一功率消耗指示符產生(i)一第一工作負載估計,且基於該第二功率預算及該第二功率消耗指示符產生(ii)一第二工作負載估計;以及基於該第一工作負載估計與該第二工作負載估計之一比較,將(i)一第一經調整功率預算提供至該第一組件且將(ii)一第二經調整功率預算提供至該第二組件。 A computing device comprising: a first component having a first power budget; a second component having a second power budget; and a power distribution controller for: obtaining (i) the first power budget A first power consumption indicator for a component, and (ii) a second power consumption indicator for the second component; generating (i) a first job based on the first power budget and the first power consumption indicator a load estimate, and generating (ii) a second workload estimate based on the second power budget and the second power consumption indicator; and based on a comparison of the first workload estimate and one of the second workload estimates, will ( i) a first adjusted power budget is provided to the first component and (ii) a second adjusted power budget is provided to the second component. 如請求項1之計算裝置,其中該第一組件包括一中央處理單元(CPU),且該第二組件包括一圖形處理單元(GPU)。 The computing device of claim 1, wherein the first component includes a central processing unit (CPU), and the second component includes a graphics processing unit (GPU). 如請求項1之計算裝置,其中該第一工作負載估計為該第一功率消耗指示符與該第一功率預算之一比率,且其中該第二工作負載估計為該第二功率消耗指示符與該第二功率預算之一比率。 The computing device of claim 1, wherein the first workload estimate is a ratio of the first power consumption indicator to the first power budget, and wherein the second workload estimate is the second power consumption indicator and the A ratio of the second power budget. 如請求項1之計算裝置,其中該功率分配控制器用以:將該第一工作負載估計及該第二工作負載估計與一臨限值進行比較;以及為產生該第一經調整功率預算及該第二經調整功率預算,基於該比較而將該第一功率預算及該第二功率預算中之一者之一部分重新分配至該第一功率預算及該第二功率預算中之另一者。 The computing device of claim 1, wherein the power distribution controller is to: compare the first workload estimate and the second workload estimate to a threshold; and to generate the first adjusted power budget and the A second adjusted power budget that reallocates a portion of the one of the first power budget and the second power budget to the other of the first power budget and the second power budget based on the comparison. 如請求項4之計算裝置,其中該功率分配控制器用以將該第 一工作負載估計及該第二工作負載估計與一下臨限值及一上臨限值進行比較。 The computing device of claim 4, wherein the power distribution controller is used to A workload estimate and the second workload estimate are compared to a lower threshold value and an upper threshold value. 如請求項4之計算裝置,其中該功率分配控制器用以:基於該第一工作負載估計與該臨限值之該比較而產生一第一工作負載狀態指示符;基於該第二工作負載估計與該臨限值之該比較而產生一第二工作負載狀態指示符;以及基於該第一工作負載狀態指示符與該第二工作負載狀態指示符之一比較而重新分配該部分。 The computing device of claim 4, wherein the power distribution controller is configured to: generate a first workload status indicator based on the comparison of the first workload estimate and the threshold; based on the second workload estimate and The comparison of the threshold values generates a second workload status indicator; and the portion is reallocated based on a comparison of the first workload status indicator with one of the second workload status indicators. 如請求項1之計算裝置,其進一步包含:一量測裝置,其用以產生該第一功率消耗指示符及該第二功率消耗指示符。 The computing device of claim 1, further comprising: a measuring device for generating the first power consumption indicator and the second power consumption indicator. 如請求項7之計算裝置,其中該量測裝置與該功率分配控制器整合。 The computing device of claim 7, wherein the measurement device is integrated with the power distribution controller. 如請求項1之計算裝置,其中該功率分配控制器用以根據一系統功率預算判定該第一功率預算及該第二功率預算。 The computing device of claim 1, wherein the power distribution controller is configured to determine the first power budget and the second power budget according to a system power budget. 一種計算裝置,其包含:一第一目標組件,其具有一第一功率預算;一第二目標組件,其具有一第二功率預算;以及一控制器,其用以:獲得(i)表示由該第一目標組件使用中之該第一功率預算之一部分的一第一工作負載狀態指示符,及(ii)表示由該第二目標組件使用中之該第二功率預算之一部分的一第二工作負載狀態指示符;以及在該第一工作負載狀態指示符不同於該第二工作負載狀態指示符時,產生(i)該第一目標組件之一第一經調整功率預算及(ii)該第二目標組件之一第二經調整功率預算。 A computing device comprising: a first target component having a first power budget; a second target component having a second power budget; and a controller to: obtain (i) represented by A first workload status indicator representing a portion of the first power budget in use by the first target device, and (ii) a second workload status indicator representing a portion of the second power budget in use by the second target device a workload status indicator; and when the first workload status indicator is different from the second workload status indicator, generating (i) a first adjusted power budget for the first target component and (ii) the A second adjusted power budget of one of the second target components. 如請求項10之計算裝置,其中該控制器用以:獲得(i)該第一目標組件之一第一工作負載估計及(ii)該第二目標組件之一第二工作負載估計;以及根據該第一工作負載估計與一臨限值之一比較選擇該第一工作負載狀態指示符;以及根據該第二工作負載估計與該臨限值之一比較選擇該第二工作負載狀態指示符。 The computing device of claim 10, wherein the controller is to: obtain (i) a first workload estimate for the first target component and (ii) a second workload estimate for the second target component; and according to the The first workload status indicator is selected by comparing the first workload estimate with one of a threshold value; and the second workload status indicator is selected based on the comparison of the second workload estimate with one of the threshold values. 如請求項11之計算裝置,其中該控制器用以:獲得(i)該第一目標組件之一第一經量測功率消耗及(ii)該第二目標組件之一第二經量測功率消耗;判定該第一工作負載估計為該第一經量測功率消耗與該第一功率預算之一比率;以及判定該第二工作負載估計為該第二經量測功率消耗與該第二功率預算之一比率。 The computing device of claim 11, wherein the controller is to: obtain (i) a first measured power consumption of the first target component and (ii) a second measured power consumption of the second target component ; determine that the first workload estimate is a ratio of the first measured power consumption to the first power budget; and determine that the second workload estimate is the second measured power consumption to the second power budget one ratio. 一種非暫時性電腦可讀媒體,其具有可由一計算裝置之一控制器執行以進行以下操作的指令之一集合:將一第一初始功率預算提供至該計算裝置之一第一組件;將一第二初始功率預算提供至該計算裝置之一第二組件;獲得(i)該第一組件之一第一工作負載估計及(ii)該第二組件之一第二工作負載估計;將該第一工作負載估計與一上臨限值及一下臨限值進行比較以產生該第一組件的一第一工作負載狀態指示符;將該第二工作負載估計與該上臨限值及該下臨限值進行比較以產生該第二組件之一第二工作負載狀態指示符; 基於該第一工作負載狀態指示符與該第二工作負載狀態指示符之間的一差,產生(i)該第一組件之一第一經調整功率預算及(ii)該第二組件之一第二經調整功率預算。 A non-transitory computer-readable medium having a set of instructions executable by a controller of a computing device to: provide a first initial power budget to a first component of the computing device; providing a second initial power budget to a second component of the computing device; obtaining (i) a first workload estimate of the first component and (ii) a second workload estimate of the second component; A workload estimate is compared with an upper threshold value and a lower threshold value to generate a first workload status indicator for the first component; the second workload estimate is compared with the upper threshold value and the lower threshold value comparing the limits to generate a second workload status indicator for one of the second components; generating (i) a first adjusted power budget for the first component and (ii) one of the second components based on a difference between the first workload status indicator and the second workload status indicator The second adjusted power budget. 如請求項13之非暫時性電腦可讀媒體,該等指令進一步可由該控制器執行以進行以下操作:判定該第一初始功率預算為一系統功率預算之一第一部分;以及判定該第二初始功率預算為該系統功率預算之一第二部分。 As in the non-transitory computer-readable medium of claim 13, the instructions are further executable by the controller to: determine that the first initial power budget is a first portion of a system power budget; and determine the second initial power budget The power budget is the second part of one of the system power budgets. 如請求項13之非暫時性電腦可讀媒體,該等指令進一步可由該控制器執行以進行以下操作:在該第一工作負載狀態指示符與該第二工作負載狀態指示符之間的該差低於一臨限值時,將該第一初始功率預算提供至該第一組件且將該第二初始功率預算提供至該第二組件。As in the non-transitory computer-readable medium of claim 13, the instructions are further executable by the controller to perform the following operations: the difference between the first workload status indicator and the second workload status indicator Below a threshold value, the first initial power budget is provided to the first component and the second initial power budget is provided to the second component.
TW110106125A 2020-07-29 2021-02-22 Power budget allocations TWI770830B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
WOPCT/US20/43955 2020-07-29
PCT/US2020/043955 WO2022025872A1 (en) 2020-07-29 2020-07-29 Power budget allocations

Publications (2)

Publication Number Publication Date
TW202205056A TW202205056A (en) 2022-02-01
TWI770830B true TWI770830B (en) 2022-07-11

Family

ID=80036021

Family Applications (1)

Application Number Title Priority Date Filing Date
TW110106125A TWI770830B (en) 2020-07-29 2021-02-22 Power budget allocations

Country Status (2)

Country Link
TW (1) TWI770830B (en)
WO (1) WO2022025872A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20250085770A1 (en) * 2023-09-08 2025-03-13 Qualcomm Incorporated POWER MONITORING AND LIMITING OF PROCESSING UNITS (PUs) IN A PROCESSOR-BASED SYSTEM TO LIMIT OVERALL POWER CONSUMPTION OF THE PROCESSOR-BASED SYSTEM

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104106053A (en) * 2012-02-08 2014-10-15 英特尔公司 Dynamic CPU GPU load balancing using power
CN111418234A (en) * 2017-09-29 2020-07-14 苹果公司 Apparatus and method for power distribution

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7562234B2 (en) * 2005-08-25 2009-07-14 Apple Inc. Methods and apparatuses for dynamic power control
US9423858B2 (en) * 2012-09-27 2016-08-23 Intel Corporation Sharing power between domains in a processor package using encoded power consumption information from a second domain to calculate an available power budget for a first domain
US9900164B2 (en) * 2015-06-10 2018-02-20 Cisco Technology, Inc. Dynamic power management

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104106053A (en) * 2012-02-08 2014-10-15 英特尔公司 Dynamic CPU GPU load balancing using power
CN111418234A (en) * 2017-09-29 2020-07-14 苹果公司 Apparatus and method for power distribution

Also Published As

Publication number Publication date
TW202205056A (en) 2022-02-01
WO2022025872A1 (en) 2022-02-03

Similar Documents

Publication Publication Date Title
CN107111350B (en) System and method for peak dynamic power management in a portable computing device
US11886878B2 (en) Distributing power shared between an accelerated processing unit and a discrete graphics processing unit
US8769316B2 (en) Dynamically allocating a power budget over multiple domains of a processor
TWI448883B (en) Power management system and method
US11157328B2 (en) Distributed processing QoS algorithm for system performance optimization under thermal constraints
TWI518586B (en) Coprocessor and system performance and power management technology
US9229503B2 (en) Thermal power budget allocation for maximum user experience
US20120216058A1 (en) System, Method and Apparatus for Energy Efficiency and Energy Conservation by Configuring Power Management Parameters During Run Time
US12524059B2 (en) Techniques for controlling computing performance for power-constrained multi-processor computing systems
US10168752B2 (en) Systems and methods for determining a sustained thermal power envelope comprising multiple heat sources
GB2603079A (en) Dynamic power/current allocation among groups of audio amplifiers and/or haptic drivers
WO2023282990A1 (en) Peripheral interface power allocation
TWI770830B (en) Power budget allocations
JP6297748B2 (en) Processor state control based on detection of producer / consumer workload serialization
JP5435133B2 (en) Information processing apparatus, information processing apparatus control method, and program
US20250123665A1 (en) Multi-port power delivery system and power allocation method thereof
TW201804654A (en) Method of thermal management and mobile device thereof
CN106020404A (en) power management method and power supply system
CN113672466B (en) Control method and device and electronic equipment
US12135528B2 (en) System and method for current flow management
US20240241557A1 (en) Control method
US20240004448A1 (en) Platform efficiency tracker

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees