Evaluating Impact of Storage on Smartphone Energy Efficiency

David T. Nguyen

LIFE IN MOBILE ERA..

1,038,000,000 SMARTPHONE USERS WORLDWIDE [IBTIMES]

27% INCREASED # SMARTPHONES SOLD ANNUALLY [IDC]

Figure Courtesy: http://www.ideas4ios.com

David T. Nguyen 2

SMARTPHONE APPS DO EVERYTHING!

850,000 APPS IN APPLE STORE 05/13 [APPLE]

800,000 APPS IN GOOGLE PLAY 05/13 [CANALYS]

145,000 APPS IN WINDOWS STORE 05/13 [CANALYS]

120,000 APPS IN BLACKBERRY WORLD 05/13 [CANALYS]

Figure Courtesy: http://aptito.com

David T. Nguyen 3

Still BIG Problem

David T. Nguyen 4

Figure Courtesy: http://cdn.cultofmac.com

Smartphone Dislikes

David T. Nguyen 5

Call Quality

Contact List Import

Excessive Dropped Calls

3G Quality

Screen Size

4G Capability

Battery Life

0% 10% 20% 30% 40%

Source: ChangeWave

OutlineIntroductionBackgroundExperimental StudyPilot SolutionEvaluation

David T. Nguyen 6

IntroductionResearching energy consumption

essential

What has been done◦Performance bottleneck in storage [Kim et al., FAST ‘12]

◦No direct study of storage – energy consumption correlation

David T. Nguyen 7

IntroductionThesis Statement

◦ Investigate impact of storage on smartphone energy efficiency

◦Explain root reasons of such impact◦Develop storage-aware energy saving

solutions

Expected Contributions ◦Better understanding of storage subsystem

and its impact on energy efficiency◦Storage-aware energy saving solutions

David T. Nguyen 8

OutlineIntroductionBackgroundExperimental StudyPilot SolutionEvaluation

David T. Nguyen 9

I/O Path

David T. Nguyen 10

Red: Nexus One default static configurations

OutlineIntroductionBackgroundExperimental StudyPilot SolutionEvaluation

David T. Nguyen 11

ApproachInvestigate impact of different

storage configurations on power levels

1. Run series of benchmarks under default configurations

2. Repeat benchmarks under different configurations

3. Compare energy consumptions David T. Nguyen 12

SetupRooted smartphone Nexus One

8 benchmarks

Monsoon Power Monitor

David T. Nguyen 13

Power Consumption: Default Config. (Queue Depth 128 / Write-back cache)

David T. Nguyen 14

Different algorithms - different power levelsNo algorithm optimal for all benchmarks Changing algorithms may save energy

Power Consumption: Queue Depth 4

David T. Nguyen 15

Shorter queue depth saves energy in most casesNot storage intensive benchmarks consume more

power due to overhead of smaller queue

Optimal ConfigurationsRun benchmarks with all

combinations of scheduling algorithms and queue depths

Record in benchmark table

David T. Nguyen 16

OutlineIntroductionBackgroundExperimental StudyPilot SolutionEvaluation

David T. Nguyen 17

Big Idea

Track phone’s run-time

I/O pattern

Match phone’s

pattern with pattern from benchmark

table

Dynamically configure parameters with optimal

savings

David T. Nguyen 18

SmartStorage Architecture

David T. Nguyen 19

GUI

David T. Nguyen 20

I/O Pattern MatchingCompare phone’s I/O pattern with

patterns from benchmark table

Matching feature:

If phone’s rate of reads/writes per second close to a benchmark from table◦ match is found

Else ◦ no match

David T. Nguyen 21

OutlineIntroductionBackgroundExperimental StudyPilot SolutionEvaluation

David T. Nguyen 22

Energy Savings: Nexus One

David T. Nguyen 23

Remaining StepsEnergy savings with different

caching policies / file systems / queue depths

Matching using machine learning

Adaptive I/O pattern recalculation

Root reasons of energy savingsDavid T. Nguyen 24

THANK YOU!

David T. Nguyen 25