Characterization of user’s behavior variations for design of replayable mobile workloads

Shruti Patil; Yeseong Kim; Kunal Korgaonkar; Ibrahim Awwal; Tajana S. Rosing

doi:10.1007/978-3-319-29003-4_4

Characterization of user’s behavior variations for design of replayable mobile workloads

Shruti Patil, Yeseong Kim, Kunal Korgaonkar, Ibrahim Awwal, Tajana S. Rosing

Department of Electrical Engineering and Computer Science

University of California at San Diego

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

8 Scopus citations

Abstract

Mobile systems leverage heterogeneous cores to deliver a desired user experience. However, how these cores cooperate in executing interactive mobile applications in the hands of a real user is unclear, preventing more realistic studies on mobile platforms. In this paper, we study how 33 users run applications on modern smartphones over a period of a month. We analyze the usage of CPUs, GPUs and associated memory operations in real user interactions, and develop microbenchmarks on an automated methodology which describes realistic and replayable test runs that statistically mimic user variations. Based on the generated test runs, we further empirically characterize memory bandwidth and power consumption of CPUs and GPUs to show the impact of user variations in the system, and identify user variationaware optimization opportunities in actual mobile application uses.

Original language	English
Title of host publication	Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers
Editors	Flora Salim, Stephan Sigg, Petteri Nurmi
Publisher	Springer Verlag
Pages	51-70
Number of pages	20
ISBN (Print)	9783319290027
DOIs	https://doi.org/10.1007/978-3-319-29003-4_4
State	Published - 2015
Event	7th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2015 - Berlin, Germany Duration: 12 Nov 2015 → 13 Nov 2015

Publication series

Name	Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
Volume	162
ISSN (Print)	1867-8211

Conference

Conference	7th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2015
Country/Territory	Germany
City	Berlin
Period	12/11/15 → 13/11/15

Bibliographical note

Publisher Copyright:
© Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2015.

Keywords

GPU usage
Heterogeneous cores
Mobile device
User variation

Access to Document

10.1007/978-3-319-29003-4_4

Cite this

Patil, S., Kim, Y., Korgaonkar, K., Awwal, I., & Rosing, T. S. (2015). Characterization of user’s behavior variations for design of replayable mobile workloads. In F. Salim, S. Sigg, & P. Nurmi (Eds.), Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers (pp. 51-70). (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST; Vol. 162). Springer Verlag. https://doi.org/10.1007/978-3-319-29003-4_4

Patil, Shruti ; Kim, Yeseong ; Korgaonkar, Kunal et al. / Characterization of user’s behavior variations for design of replayable mobile workloads. Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers. editor / Flora Salim ; Stephan Sigg ; Petteri Nurmi. Springer Verlag, 2015. pp. 51-70 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST).

@inproceedings{1801dd1b15274a9cbe7d8bfe376c4649,

title = "Characterization of user{\textquoteright}s behavior variations for design of replayable mobile workloads",

abstract = "Mobile systems leverage heterogeneous cores to deliver a desired user experience. However, how these cores cooperate in executing interactive mobile applications in the hands of a real user is unclear, preventing more realistic studies on mobile platforms. In this paper, we study how 33 users run applications on modern smartphones over a period of a month. We analyze the usage of CPUs, GPUs and associated memory operations in real user interactions, and develop microbenchmarks on an automated methodology which describes realistic and replayable test runs that statistically mimic user variations. Based on the generated test runs, we further empirically characterize memory bandwidth and power consumption of CPUs and GPUs to show the impact of user variations in the system, and identify user variationaware optimization opportunities in actual mobile application uses.",

keywords = "GPU usage, Heterogeneous cores, Mobile device, User variation",

author = "Shruti Patil and Yeseong Kim and Kunal Korgaonkar and Ibrahim Awwal and Rosing, {Tajana S.}",

note = "Publisher Copyright: {\textcopyright} Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2015.; 7th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2015 ; Conference date: 12-11-2015 Through 13-11-2015",

year = "2015",

doi = "10.1007/978-3-319-29003-4_4",

language = "English",

isbn = "9783319290027",

series = "Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST",

publisher = "Springer Verlag",

pages = "51--70",

editor = "Flora Salim and Stephan Sigg and Petteri Nurmi",

booktitle = "Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers",

}

Patil, S, Kim, Y, Korgaonkar, K, Awwal, I & Rosing, TS 2015, Characterization of user’s behavior variations for design of replayable mobile workloads. in F Salim, S Sigg & P Nurmi (eds), Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers. Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST, vol. 162, Springer Verlag, pp. 51-70, 7th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2015, Berlin, Germany, 12/11/15. https://doi.org/10.1007/978-3-319-29003-4_4

Characterization of user’s behavior variations for design of replayable mobile workloads. / Patil, Shruti; Kim, Yeseong; Korgaonkar, Kunal et al.
Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers. ed. / Flora Salim; Stephan Sigg; Petteri Nurmi. Springer Verlag, 2015. p. 51-70 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST; Vol. 162).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Characterization of user’s behavior variations for design of replayable mobile workloads

AU - Patil, Shruti

AU - Kim, Yeseong

AU - Korgaonkar, Kunal

AU - Awwal, Ibrahim

AU - Rosing, Tajana S.

N1 - Publisher Copyright: © Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2015.

PY - 2015

Y1 - 2015

N2 - Mobile systems leverage heterogeneous cores to deliver a desired user experience. However, how these cores cooperate in executing interactive mobile applications in the hands of a real user is unclear, preventing more realistic studies on mobile platforms. In this paper, we study how 33 users run applications on modern smartphones over a period of a month. We analyze the usage of CPUs, GPUs and associated memory operations in real user interactions, and develop microbenchmarks on an automated methodology which describes realistic and replayable test runs that statistically mimic user variations. Based on the generated test runs, we further empirically characterize memory bandwidth and power consumption of CPUs and GPUs to show the impact of user variations in the system, and identify user variationaware optimization opportunities in actual mobile application uses.

AB - Mobile systems leverage heterogeneous cores to deliver a desired user experience. However, how these cores cooperate in executing interactive mobile applications in the hands of a real user is unclear, preventing more realistic studies on mobile platforms. In this paper, we study how 33 users run applications on modern smartphones over a period of a month. We analyze the usage of CPUs, GPUs and associated memory operations in real user interactions, and develop microbenchmarks on an automated methodology which describes realistic and replayable test runs that statistically mimic user variations. Based on the generated test runs, we further empirically characterize memory bandwidth and power consumption of CPUs and GPUs to show the impact of user variations in the system, and identify user variationaware optimization opportunities in actual mobile application uses.

KW - GPU usage

KW - Heterogeneous cores

KW - Mobile device

KW - User variation

UR - http://www.scopus.com/inward/record.url?scp=84956635522&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-29003-4_4

DO - 10.1007/978-3-319-29003-4_4

M3 - Conference contribution

AN - SCOPUS:84956635522

SN - 9783319290027

T3 - Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST

SP - 51

EP - 70

BT - Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers

A2 - Salim, Flora

A2 - Sigg, Stephan

A2 - Nurmi, Petteri

PB - Springer Verlag

T2 - 7th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2015

Y2 - 12 November 2015 through 13 November 2015

ER -

Patil S, Kim Y, Korgaonkar K, Awwal I, Rosing TS. Characterization of user’s behavior variations for design of replayable mobile workloads. In Salim F, Sigg S, Nurmi P, editors, Mobile Computing, Applications, and Services - 7th International Conference, MobiCASE 2015, Revised Selected Papers. Springer Verlag. 2015. p. 51-70. (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST). doi: 10.1007/978-3-319-29003-4_4

Characterization of user’s behavior variations for design of replayable mobile workloads

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this