More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design

Jiacheng Huang; Yunmo Zhang; Junqiao Qiu; Yu Liang; Rachata Ausavarungnirun; Qingan Li; Chun Jason Xue

doi:10.1145/3620666.3651377

More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design

Jiacheng Huang, Yunmo Zhang, Junqiao Qiu, Yu Liang, Rachata Ausavarungnirun, Qingan Li^*, Chun Jason Xue

^*Corresponding author for this work

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

7 Citations (Scopus)

Abstract

Faster app launching is crucial for the user experience on mobile devices. Apps launched from a background cached state, called hot-launching, have much better performance than apps launched from scratch. To increase the number of hot-launches, leading mobile vendors now cache more apps in the background by enabling swap. Recent work also proposed reducing the Java heap to increase the number of cached apps. However, this paper found that existing methods deteriorate app hot-launch performance while increasing the number of cached apps. To simultaneously improve the number of cached apps and hot-launch performance, this paper proposes Fleet, a foreground/background-aware GC-swap co-design framework. To enhance app-caching capacity, Fleet limits the tracing range of GC to background objects only, avoiding touching long-lifetime foreground objects. To improve hot-launch performance, Fleet identifies objects that will be accessed during the next hot-launch and uses runtime information to guide the swap scheme in the OS. In addition, Fleet aggregates small objects with similar access patterns into the same pages to improve swap efficiency. We implemented Fleet in AOSP and evaluated its performance with different types of apps. Experimental results show that Fleet achieves a 1.59× faster hot-launch time and caches 1.21× more apps than Android. © 2024 Copyright held by the owner/author(s).

Original language	English
Title of host publication	ASPLOS '24
Subtitle of host publication	Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3
Place of Publication	New York, NY
Publisher	Association for Computing Machinery
Pages	654-670
Number of pages	17
ISBN (Print)	979-8-4007-0386-7
DOIs	https://doi.org/10.1145/3620666.3651377
Publication status	Published - 2024
Event	29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2024) - San Diego, United States Duration: 27 Apr 2024 → 1 May 2024 https://www.asplos-conference.org/asplos2024/

Conference

Conference	29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2024)
Abbreviated title	ASPLOS ’24
Country/Territory	United States
City	San Diego
Period	27/04/24 → 1/05/24
Internet address	https://www.asplos-conference.org/asplos2024/

Bibliographical note

Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

Funding

We are grateful to our shepherd and the anonymous reviewers for their constructive comments and suggestions. The work was supported in part by the Research Grants Council of the Hong Kong Special Administrative Region, China (No. CityU 11209122), the Thailand MHESI Research Grant for New Scholars (RGNS 64-091), the National Key Research and Development Program of China (No. 2022YFB3104502), the State Key Laboratory of Computer Architecture (ICT, CAS) under Grant CARCH A20211, the Wuhan Science and Technology Joint Project for Building a Strong Transportation Country (No. 2023-2-7), the CCF-Phytium Fund, and the National Science Foundation (NSF) Grant (No. 2105006).

Research Keywords

Garbage Collection
Memory Management
Mobile Systems
Operating Systems

Access Output

10.1145/3620666.3651377

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Huang, J., Zhang, Y., Qiu, J., Liang, Y., Ausavarungnirun, R., Li, Q., & Xue, C. J. (2024). More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design. In ASPLOS '24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3 (pp. 654-670). Association for Computing Machinery. https://doi.org/10.1145/3620666.3651377

@inproceedings{dbb78e30048943e8a03f0c087e189106,

title = "More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design",

abstract = "Faster app launching is crucial for the user experience on mobile devices. Apps launched from a background cached state, called hot-launching, have much better performance than apps launched from scratch. To increase the number of hot-launches, leading mobile vendors now cache more apps in the background by enabling swap. Recent work also proposed reducing the Java heap to increase the number of cached apps. However, this paper found that existing methods deteriorate app hot-launch performance while increasing the number of cached apps. To simultaneously improve the number of cached apps and hot-launch performance, this paper proposes Fleet, a foreground/background-aware GC-swap co-design framework. To enhance app-caching capacity, Fleet limits the tracing range of GC to background objects only, avoiding touching long-lifetime foreground objects. To improve hot-launch performance, Fleet identifies objects that will be accessed during the next hot-launch and uses runtime information to guide the swap scheme in the OS. In addition, Fleet aggregates small objects with similar access patterns into the same pages to improve swap efficiency. We implemented Fleet in AOSP and evaluated its performance with different types of apps. Experimental results show that Fleet achieves a 1.59× faster hot-launch time and caches 1.21× more apps than Android. {\textcopyright} 2024 Copyright held by the owner/author(s).",

keywords = "Garbage Collection, Memory Management, Mobile Systems, Operating Systems",

author = "Jiacheng Huang and Yunmo Zhang and Junqiao Qiu and Yu Liang and Rachata Ausavarungnirun and Qingan Li and Xue, {Chun Jason}",

note = "Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).; 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2024), ASPLOS {\textquoteright}24 ; Conference date: 27-04-2024 Through 01-05-2024",

year = "2024",

doi = "10.1145/3620666.3651377",

language = "English",

isbn = "979-8-4007-0386-7",

pages = "654--670",

booktitle = "ASPLOS '24",

publisher = "Association for Computing Machinery",

address = "United States",

url = "https://www.asplos-conference.org/asplos2024/",

}

Huang, J , Zhang, Y , Qiu, J, Liang, Y, Ausavarungnirun, R, Li, Q & Xue, CJ 2024, More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design. in ASPLOS '24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3. Association for Computing Machinery, New York, NY, pp. 654-670, 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2024), San Diego, California, United States, 27/04/24. https://doi.org/10.1145/3620666.3651377

More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design. / Huang, Jiacheng ; Zhang, Yunmo ; Qiu, Junqiao et al.
ASPLOS '24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3. New York, NY: Association for Computing Machinery, 2024. p. 654-670.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design

AU - Huang, Jiacheng

AU - Zhang, Yunmo

AU - Qiu, Junqiao

AU - Liang, Yu

AU - Ausavarungnirun, Rachata

AU - Li, Qingan

AU - Xue, Chun Jason

N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

PY - 2024

Y1 - 2024

N2 - Faster app launching is crucial for the user experience on mobile devices. Apps launched from a background cached state, called hot-launching, have much better performance than apps launched from scratch. To increase the number of hot-launches, leading mobile vendors now cache more apps in the background by enabling swap. Recent work also proposed reducing the Java heap to increase the number of cached apps. However, this paper found that existing methods deteriorate app hot-launch performance while increasing the number of cached apps. To simultaneously improve the number of cached apps and hot-launch performance, this paper proposes Fleet, a foreground/background-aware GC-swap co-design framework. To enhance app-caching capacity, Fleet limits the tracing range of GC to background objects only, avoiding touching long-lifetime foreground objects. To improve hot-launch performance, Fleet identifies objects that will be accessed during the next hot-launch and uses runtime information to guide the swap scheme in the OS. In addition, Fleet aggregates small objects with similar access patterns into the same pages to improve swap efficiency. We implemented Fleet in AOSP and evaluated its performance with different types of apps. Experimental results show that Fleet achieves a 1.59× faster hot-launch time and caches 1.21× more apps than Android. © 2024 Copyright held by the owner/author(s).

AB - Faster app launching is crucial for the user experience on mobile devices. Apps launched from a background cached state, called hot-launching, have much better performance than apps launched from scratch. To increase the number of hot-launches, leading mobile vendors now cache more apps in the background by enabling swap. Recent work also proposed reducing the Java heap to increase the number of cached apps. However, this paper found that existing methods deteriorate app hot-launch performance while increasing the number of cached apps. To simultaneously improve the number of cached apps and hot-launch performance, this paper proposes Fleet, a foreground/background-aware GC-swap co-design framework. To enhance app-caching capacity, Fleet limits the tracing range of GC to background objects only, avoiding touching long-lifetime foreground objects. To improve hot-launch performance, Fleet identifies objects that will be accessed during the next hot-launch and uses runtime information to guide the swap scheme in the OS. In addition, Fleet aggregates small objects with similar access patterns into the same pages to improve swap efficiency. We implemented Fleet in AOSP and evaluated its performance with different types of apps. Experimental results show that Fleet achieves a 1.59× faster hot-launch time and caches 1.21× more apps than Android. © 2024 Copyright held by the owner/author(s).

KW - Garbage Collection

KW - Memory Management

KW - Mobile Systems

KW - Operating Systems

UR - http://www.scopus.com/inward/record.url?scp=85192183396&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85192183396&origin=recordpage

U2 - 10.1145/3620666.3651377

DO - 10.1145/3620666.3651377

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 979-8-4007-0386-7

SP - 654

EP - 670

BT - ASPLOS '24

PB - Association for Computing Machinery

CY - New York, NY

T2 - 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2024)

Y2 - 27 April 2024 through 1 May 2024

ER -

Huang J , Zhang Y , Qiu J, Liang Y, Ausavarungnirun R, Li Q et al. More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design. In ASPLOS '24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3. New York, NY: Association for Computing Machinery. 2024. p. 654-670 Epub 2024 Apr 27. doi: 10.1145/3620666.3651377

More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design

Abstract

Conference

Bibliographical note

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Towards Unified-storage-memory-enabled Mobile Devices

Cite this

More Apps, Faster Hot-Launch on Mobile Devices via Fore/Background-aware GC-Swap Co-design

Abstract

Conference

Bibliographical note

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Towards Unified-storage-memory-enabled Mobile Devices

Cite this