SlideShare a Scribd company logo
Power Management in Mobile Phones Need and Techniques
Agenda What is Power Management? Why Power Management? Power Management Techniques HW support .. Role of Applications .. Power Management and Media use cases
What is power Management? Technique for effective use of energy Desired performance at minimal power consumption Automated without user intervention A little help from Applications is always welcome!!
Why Power Management? Applications and usage have become sophisticated Requiring More complicated chips Freq of operation directly proportional to power consumed Main chip is not the only consumer Bigger color touch screen consumes more power More RAM requires more power Each new peripheral like WLAN and BT requires power But resource & energy constrained! Breakthrough in battery technology is not at same pace of the  Most phones will have about 1000 mAh battery Let us see how long this battery last? Only 80% of the battery is usable  At 500mA it will last less then two hours Some figures 0.6mW/MHz (0.13µm, 1.2V) including cache controllers for ARM11 This means processor only  consumes .5mA/MHz or ~300mA at full 600MHz
Why Power Management?.. All capabilities are not always used Most applications do not require 100% MIPS Most HW accelerators or component are not used often Most peripherals not used often There is an Idle time between active use How to handle such cases Its like complicated switch board with multiple switches Some On/Off some variable dimmer Effective management of this required
Power Management Technique Simple energy management directives: Power off component when not in use (USB) Slow down whenever possible (processor) Efficient schemes: Dynamic power management (DPM) system switches dynamically between high- and low-consumption system power modes based on activity Dynamic voltage and frequency scaling (DVFS) DVFS is a technique through which the idle time of the system is minimized. It uses the optimal operating frequency and voltage to allow a task to be performed in the required amount of time. This reduces the active power consumption (power consumed while executing a task) of the device while still meeting task requirements.
DPM
Power Management Technique… Decision: DVFS or DPM? DPM - High latency for Power Off to Power On  Voltage stabilization time Clock stabilization time Dependencies chain make it even more DVFS also require clock stabilization There are other techniques like cutting clock - low latency Not that useful as in 45 nm considerable Power is lost even when not switching – leakage current ARM in sleep will also consume power But until we have sufficient sleep time we cannot switch off
HW Support .. Support Multiple Voltage Levels (Voltage domain) Power = V*I , lower voltage means lower power Chip is divided  in multiple Voltage domains Voltage level controlled by SW “DVFS” Support Multiple Power Domains Having switches for controlling power to sections of chip But the number and dependencies is unmanageable HW helps here by managing dependencies Support for  Clock Gating (Clock domain) Switching consumes power  Some power can be saved by cutting clock This is quickest thing to do Here also number of clocks and there dependencies are complex  HW helps here in managing dependencies
Role of Applications.. Allow processor to go to sleep For longer sleep duration DPM kicks in Avoid periodic polling This impacts the sleep duration and type Unnecessary wakeups increases power consumptions Instead waiting for events/interrupts If required and possible try using deferrable timers Some OS provides  it e.g. Symbian  Optimize computation to reduce MIPS For higher MIPS, HW chooses high freq/voltage Utilize HWA as much as possible Use OEM supplied libraries and routines for complex algorithms E.g. Video decode, these are profiled and Tuned for optimal power consumption
Power Management and Media UC Media Play (MP3) Scenario Classification: Final frame rendering at fixed rate Much slower then time it may take to decode a frame Recommendation Fetch and Decode frames in chunk While decoded frames are rendered only audio peripheral, system DMA , memory etc. will be powered and not processor Impact High power consuming resource remain idle for a large chunk of time With single frame decode total idle for processor is X but it wakes-up frequently inhabiting deeper sleeps With multiple frame total idle time still remains X but system wakes-up less frequently So while Media is playing processor does go to sleep App should still refrain from periodic polling
Thanks
Power Management Value chain(Backup)

More Related Content

Similar to Power_Management_In_Mobiles

11. Computer Systems Hardware 1
11. Computer Systems   Hardware 111. Computer Systems   Hardware 1
11. Computer Systems Hardware 1
New Era University
 
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
Edge AI and Vision Alliance
 
Literature survey presentation
Literature survey presentationLiterature survey presentation
Literature survey presentation
Karthik Iyr
 
Modern processor art
Modern processor artModern processor art
Modern processor art
waqasjadoon11
 

Similar to Power_Management_In_Mobiles (20)

Os Vandeven
Os VandevenOs Vandeven
Os Vandeven
 
11. Computer Systems Hardware 1
11. Computer Systems   Hardware 111. Computer Systems   Hardware 1
11. Computer Systems Hardware 1
 
Embedded systems-unit-1
Embedded systems-unit-1Embedded systems-unit-1
Embedded systems-unit-1
 
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
“Battery-powered Edge AI Sensing: A Case Study Implementing Low-power, Always...
 
Dynamic Voltage and Frequency Scaling
Dynamic Voltage and Frequency ScalingDynamic Voltage and Frequency Scaling
Dynamic Voltage and Frequency Scaling
 
Green IT
Green ITGreen IT
Green IT
 
Energy Efficiency in Large Scale Systems
Energy Efficiency in Large Scale SystemsEnergy Efficiency in Large Scale Systems
Energy Efficiency in Large Scale Systems
 
Literature survey presentation
Literature survey presentationLiterature survey presentation
Literature survey presentation
 
Power Management in Embedded Systems
Power Management in Embedded Systems Power Management in Embedded Systems
Power Management in Embedded Systems
 
computer organization and architecture notes
computer organization and architecture notescomputer organization and architecture notes
computer organization and architecture notes
 
Modern processor art
Modern processor artModern processor art
Modern processor art
 
processor struct
processor structprocessor struct
processor struct
 
Modern processor art
Modern processor artModern processor art
Modern processor art
 
Danish presentation
Danish presentationDanish presentation
Danish presentation
 
06. System unit of computer
06. System unit of computer06. System unit of computer
06. System unit of computer
 
Chapter 4
Chapter 4Chapter 4
Chapter 4
 
Reduce system energy by smart CPU management
Reduce system energy by smart CPU managementReduce system energy by smart CPU management
Reduce system energy by smart CPU management
 
Computers8 Ch4 3
Computers8 Ch4 3Computers8 Ch4 3
Computers8 Ch4 3
 
Lecture1
Lecture1Lecture1
Lecture1
 
Unit-I_ES.pdf
Unit-I_ES.pdfUnit-I_ES.pdf
Unit-I_ES.pdf
 

Recently uploaded

Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo DiehlFuture Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
Peter Udo Diehl
 
Essentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with ParametersEssentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with Parameters
Safe Software
 

Recently uploaded (20)

Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualitySoftware Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
 
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...
 
Knowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and backKnowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and back
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
 
Assuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesAssuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyes
 
To Graph or Not to Graph Knowledge Graph Architectures and LLMs
To Graph or Not to Graph Knowledge Graph Architectures and LLMsTo Graph or Not to Graph Knowledge Graph Architectures and LLMs
To Graph or Not to Graph Knowledge Graph Architectures and LLMs
 
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
 
Ransomware Mallox [EN].pdf
Ransomware         Mallox       [EN].pdfRansomware         Mallox       [EN].pdf
Ransomware Mallox [EN].pdf
 
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
 
Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo DiehlFuture Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
Future Visions: Predictions to Guide and Time Tech Innovation, Peter Udo Diehl
 
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
 
Accelerate your Kubernetes clusters with Varnish Caching
Accelerate your Kubernetes clusters with Varnish CachingAccelerate your Kubernetes clusters with Varnish Caching
Accelerate your Kubernetes clusters with Varnish Caching
 
PHP Frameworks: I want to break free (IPC Berlin 2024)
PHP Frameworks: I want to break free (IPC Berlin 2024)PHP Frameworks: I want to break free (IPC Berlin 2024)
PHP Frameworks: I want to break free (IPC Berlin 2024)
 
Connector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a buttonConnector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a button
 
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
 
Essentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with ParametersEssentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with Parameters
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
 
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdfFIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
 
Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 

Power_Management_In_Mobiles

  • 1. Power Management in Mobile Phones Need and Techniques
  • 2. Agenda What is Power Management? Why Power Management? Power Management Techniques HW support .. Role of Applications .. Power Management and Media use cases
  • 3. What is power Management? Technique for effective use of energy Desired performance at minimal power consumption Automated without user intervention A little help from Applications is always welcome!!
  • 4. Why Power Management? Applications and usage have become sophisticated Requiring More complicated chips Freq of operation directly proportional to power consumed Main chip is not the only consumer Bigger color touch screen consumes more power More RAM requires more power Each new peripheral like WLAN and BT requires power But resource & energy constrained! Breakthrough in battery technology is not at same pace of the Most phones will have about 1000 mAh battery Let us see how long this battery last? Only 80% of the battery is usable At 500mA it will last less then two hours Some figures 0.6mW/MHz (0.13µm, 1.2V) including cache controllers for ARM11 This means processor only consumes .5mA/MHz or ~300mA at full 600MHz
  • 5. Why Power Management?.. All capabilities are not always used Most applications do not require 100% MIPS Most HW accelerators or component are not used often Most peripherals not used often There is an Idle time between active use How to handle such cases Its like complicated switch board with multiple switches Some On/Off some variable dimmer Effective management of this required
  • 6. Power Management Technique Simple energy management directives: Power off component when not in use (USB) Slow down whenever possible (processor) Efficient schemes: Dynamic power management (DPM) system switches dynamically between high- and low-consumption system power modes based on activity Dynamic voltage and frequency scaling (DVFS) DVFS is a technique through which the idle time of the system is minimized. It uses the optimal operating frequency and voltage to allow a task to be performed in the required amount of time. This reduces the active power consumption (power consumed while executing a task) of the device while still meeting task requirements.
  • 7. DPM
  • 8. Power Management Technique… Decision: DVFS or DPM? DPM - High latency for Power Off to Power On Voltage stabilization time Clock stabilization time Dependencies chain make it even more DVFS also require clock stabilization There are other techniques like cutting clock - low latency Not that useful as in 45 nm considerable Power is lost even when not switching – leakage current ARM in sleep will also consume power But until we have sufficient sleep time we cannot switch off
  • 9. HW Support .. Support Multiple Voltage Levels (Voltage domain) Power = V*I , lower voltage means lower power Chip is divided in multiple Voltage domains Voltage level controlled by SW “DVFS” Support Multiple Power Domains Having switches for controlling power to sections of chip But the number and dependencies is unmanageable HW helps here by managing dependencies Support for Clock Gating (Clock domain) Switching consumes power Some power can be saved by cutting clock This is quickest thing to do Here also number of clocks and there dependencies are complex HW helps here in managing dependencies
  • 10. Role of Applications.. Allow processor to go to sleep For longer sleep duration DPM kicks in Avoid periodic polling This impacts the sleep duration and type Unnecessary wakeups increases power consumptions Instead waiting for events/interrupts If required and possible try using deferrable timers Some OS provides it e.g. Symbian Optimize computation to reduce MIPS For higher MIPS, HW chooses high freq/voltage Utilize HWA as much as possible Use OEM supplied libraries and routines for complex algorithms E.g. Video decode, these are profiled and Tuned for optimal power consumption
  • 11. Power Management and Media UC Media Play (MP3) Scenario Classification: Final frame rendering at fixed rate Much slower then time it may take to decode a frame Recommendation Fetch and Decode frames in chunk While decoded frames are rendered only audio peripheral, system DMA , memory etc. will be powered and not processor Impact High power consuming resource remain idle for a large chunk of time With single frame decode total idle for processor is X but it wakes-up frequently inhabiting deeper sleeps With multiple frame total idle time still remains X but system wakes-up less frequently So while Media is playing processor does go to sleep App should still refrain from periodic polling
  • 13. Power Management Value chain(Backup)