Skip to main navigation Skip to search Skip to main content

Reducing Mobile Application Cold-Start Latency via Launch-Critical Data Identification

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

Abstract

Application cold start is often slowed by page cache misses when launch-critical data must be fetched from flash storage under limited memory capacity. In this paper, we analyze application launch behavior to identify the critical data that most affects launch latency, and refine launch-Time measurement to include both activity launch and interface display. Based on this analysis, we propose FLCD, a lightweight mechanism that identifies and retains launch-critical data in memory instead of prefetching all launch-related data. Evaluation on Android smartphones shows that FLCD reduces application launch time by 25% on average compared to the default Android launch mechanism, with very small memory overhead. © 2026 IEEE.
Original languageEnglish
Title of host publication2026 IEEE 15th Non-Volatile Memory Systems and Applications Symposium (NVMSA)
PublisherIEEE
Number of pages6
ISBN (Electronic)979-8-3195-4843-6
ISBN (Print)979-8-3195-4844-3
DOIs
Publication statusPublished - 2026
Event15th IEEE Non-Volatile Memory Systems and Applications Symposium (NVMSA 2026) - Qingdao, China
Duration: 11 Aug 2026 → 13 Aug 2026

Publication series

NameProceedings - IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA
ISSN (Print)2575-2561
ISSN (Electronic)2575-257X

Conference

Conference15th IEEE Non-Volatile Memory Systems and Applications Symposium (NVMSA 2026)
PlaceChina
CityQingdao
Period11/08/26 → 13/08/26

Research Keywords

  • flash storage
  • launch-critical data
  • memory management
  • mobile devices
  • page cache

Fingerprint

Dive into the research topics of 'Reducing Mobile Application Cold-Start Latency via Launch-Critical Data Identification'. Together they form a unique fingerprint.

Cite this