Mobile Applications Alberto Garca Estvez University of Alcala - - PowerPoint PPT Presentation

▶
mobile applications
SMART_READER_LITE
LIVE PREVIEW

Mobile Applications Alberto Garca Estvez University of Alcala - - PowerPoint PPT Presentation

Apr 02, 2024 712 likes •1.09k views

Geo-location-aware Emulations for Performance Evaluation of Mobile Applications Alberto Garca Estvez University of Alcala Niklas Carlsson Linkping University @ WONS 2014 , Obergurgl, Austria, April 2014 Customized service Access to

slide-1
SLIDE 1

@ WONS 2014, Obergurgl, Austria, April 2014

Geo-location-aware Emulations for Performance Evaluation of Mobile Applications

Alberto García Estévez University of Alcala Niklas Carlsson Linköping University

slide-2
SLIDE 2
  • Access to Internet everywhere
  • Wireless connectivity
  • Increasingly mobile users
  • Smart phones and tablets
  • Connected (close to) all the time
  • Powerful customized applications
  • Location-aware app
  • Customized services based on location

Customized service

2

slide-3
SLIDE 3
  • New emerging location-based services

and applications for mobile users

  • Many alternative implementations
  • Need fair evaluation methodology

Evaluation methodology

3

slide-4
SLIDE 4
  • New emerging location-based services

and applications for mobile users

  • Many alternative implementations
  • Need fair evaluation methodology

Evaluation methodology

4

slide-5
SLIDE 5
  • New emerging location-based services

and applications for mobile users

  • Many alternative implementations
  • Need fair evaluation methodology

Evaluation methodology

5

slide-6
SLIDE 6
  • New emerging location-based services

and applications for mobile users

  • Many alternative implementations
  • Need fair evaluation methodology

Evaluation methodology

6

  • Fair head-to-head comparisons …

…. under realistic scenarios

  • Repeatable experiments
  • Quick and low price
slide-7
SLIDE 7

7

Approach Comment Choice Field tests Modeling Simulations Emulation

Evaluation Methodology

slide-8
SLIDE 8

8

Approach Comment Choice Field tests Expensive and does not allow repeatable experiments Modeling Simulations Emulation

Evaluation Methodology

slide-9
SLIDE 9

9

Approach Comment Choice Field tests Expensive and does not allow repeatable experiments Modeling Simulations Difficult to ensure that abstraction matches reality Emulation

Evaluation Methodology

slide-10
SLIDE 10

10

Approach Comment Choice Field tests Expensive and does not allow repeatable experiments Modeling Simulations Difficult to ensure that abstraction matches reality Emulation Relatively cheap, real hardware, but we still need methodology for repeatable location- based evaluation …

Evaluation Methodology

slide-11
SLIDE 11

11

  • Repeatable experiments: Allow head-to-head comparison
  • Quick and low price: Can be done in-house
  • Realistic scenarios: Use of real mobility patterns and network conditions

… develop simple methodology that allow …

Approach Comment Choice Field tests Expensive and does not allow repeatable experiments Modeling Simulations Difficult to ensure that abstraction matches reality Emulation Relatively cheap, real hardware, but we still need methodology for repeatable location- based evaluation …

Evaluation Methodology

slide-12
SLIDE 12
  • Location-aware download scheduler

based on notification service

  • Google Cloud Messaging (GCM)
  • Mobile app
  • HTC wildfire with Android
  • Wi-Fi and location service (GPS and

network)

  • Application server
  • PHP + MySql
  • Notifications, network conditions

Example application

12

slide-13
SLIDE 13

13

Example application

Registration Notifications

slide-14
SLIDE 14

14

Example application

Registration Notifications

slide-15
SLIDE 15

15

Example application

1.

The server sends a notification to GCM

2.

GCM notifies the mobile that an update is available

3.

The mobile requests the update

4.

The server sends the update

slide-16
SLIDE 16

16

Example application

1.

The server sends a notification to GCM

2.

GCM notifies the mobile that an update is available

3.

The mobile requests the update [** geoSmart Scheduler** ]

4.

The server sends the update

slide-17
SLIDE 17

GeoSmart Scheduler

  • - Design and Proof-of-concept Implementation

17

Performance Network Map + Smart Scheduler

slide-18
SLIDE 18

GeoSmart Scheduler

  • - Design and Proof-of-concept Implementation

18

Performance Network Map + Smart Scheduler

slide-19
SLIDE 19

Throughput-location pairs

  • HTTP throughput prediction

1. Passively measure throughput when data is downloaded 2. Update prediction using EWMA

  • UTM location:

1. Obtain location in latitude/longitude when data is downloaded 2. Convert location to UTM coordinates

Performance Network Map

19

slide-20
SLIDE 20

GeoSmart Scheduler

  • - Design and Proof-of-concept Implementation

20

Performance Network Map + Smart Scheduler

slide-21
SLIDE 21

21

GeoSmart Scheduler

Basic implementation

 FIFO Notifications

queue using

 Threshold based on

average path throughput

slide-22
SLIDE 22

TRACE-BASED EMULATION EVALUATION

Evaluation and results

22

slide-23
SLIDE 23

Trace-driven emulation

23

 Client location and bandwidth conditions  Traces obtained from dataset of real measurements  E.g., commuter traces: bus, ferry, car, train, etc.  (i) Timestamp, (ii) Latitude/longitude, and (iii) bandwidth  Location mocking using Android API features  Create test location service  Network conditions emulated with Dummynet  Server-driven workload  Traces collected using Twitter API  E.g., rate of 3 to 12 notifications per minute  (i) time stamp and (ii) unique ID

slide-24
SLIDE 24

Trace-driven emulation

24

 Client location and bandwidth conditions  Traces obtained from dataset of real measurements  E.g., commuter traces: bus, ferry, car, train, etc.  (i) Timestamp, (ii) Latitude/longitude, and (iii) Bandwidth  Location mocking using Android API features  Create test location service  Network conditions emulated with Dummynet  Server-driven workload  Traces collected using Twitter API  E.g., rate of 3 to 12 notifications per minute  (i) time stamp and (ii) unique ID

slide-25
SLIDE 25

Bandwidth, location, and and workload traces

25

Bus scenario

  • H. Riiser, P. Vigmostad, C. Griwodz, and P. Halvorsen, “Commute path bandwidth traces

from 3g networks: Analysis and applications,” in Proc. ACM MMSys, Feb/Mar. 2013.

slide-26
SLIDE 26

Bandwidth, location, and and workload traces

26

Bus scenario

(a) Bus (b) Ferry (c) Metro (d) Tram

slide-27
SLIDE 27

Bandwidth, location, and and workload traces

27

Bus scenario #topicX #topicY Notification traces … Time 1 2 3 4 5 6 78 9

slide-28
SLIDE 28

Naive download speeds

28

Sample file size 100KB

Bus scenario Ferry scenario

slide-29
SLIDE 29

GeoSmart Scheduler Results

  • Example measure: Average download time
  • Three (3) alternative approaches (or grid sizes)
  • Four (4) alternative file sizes

Bus scenario Ferry scenario

29

slide-30
SLIDE 30

GeoSmart Scheduler Results

  • Example measure: Average download time
  • Three (3) alternative approaches (or grid sizes)
  • Four (4) alternative file sizes

Bus scenario Ferry scenario

30

slide-31
SLIDE 31

GeoSmart Scheduler Results

  • Example measure: Average download time
  • Three (3) alternative approaches (or grid sizes)
  • Four (4) alternative file sizes

Bus scenario Ferry scenario

31

slide-32
SLIDE 32

GeoSmart Scheduler Results

Bus scenario Ferry scenario

32

  • Relatively small improvements (e.g, 10-20%)
slide-33
SLIDE 33

GeoSmart Scheduler Results

Bus scenario Ferry scenario

33

  • Relatively small improvements
  • Better improvements in scenarios with significant

location differences in network performance

slide-34
SLIDE 34
  • Our emulation framework provides fair-head-to-head

protocol/service comparisons

  • Real hardware and realistic mobile scenarios
  • Repeatable experiments
  • Relatively low cost
  • Regards to our proof-of-concept implementation
  • GeoSmart scheduler perform better in scenarios with significant location

differences in network performance

  • Limited accuracy of EWMA estimator for HTTP throughput
  • Choose correct resolution is important
  • Future work will consider
  • Higher order stochastic models for estimation, adaptive map resolution

(e.g., based on speed of user) with richer information (e.g., based on network data technology)

Conclusions

34

slide-35
SLIDE 35
  • Our emulation framework provides fair-head-to-head

protocol/service comparisons

  • Real hardware and realistic mobile scenarios
  • Repeatable experiments
  • Relatively low cost
  • Regards to our proof-of-concept implementation
  • GeoSmart scheduler perform better in scenarios with significant location

differences in network performance

  • Limited accuracy of EWMA estimator for HTTP throughput
  • Choose correct resolution is important
  • Future work will consider
  • Higher order stochastic models for estimation, adaptive map resolution

(e.g., based on speed of user) with richer information (e.g., based on network data technology)

Conclusions

35

slide-36
SLIDE 36
  • Our emulation framework provides fair-head-to-head

protocol/service comparisons

  • Real hardware and realistic mobile scenarios
  • Repeatable experiments
  • Relatively low cost
  • Regards to our proof-of-concept implementation
  • GeoSmart scheduler perform better in scenarios with significant location

differences in network performance

  • Limited accuracy of EWMA estimator for HTTP throughput
  • Choose correct resolution is important
  • Future work will consider
  • Higher order stochastic models for estimation, adaptive map resolution

(e.g., based on speed of user) with richer information (e.g., based on network data technology)

Conclusions

36

slide-37
SLIDE 37

www.liu.se

Alberto García Estévez (UA) Niklas Carlsson (LiU)

Geo-location-aware Emulations for Performance Evaluation of Mobile Applications

Software: www.ida.liu.se/~nikca/papers/wons14.html