Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AppFlow: Memory Scheduling for Cold Launch of Large Apps on Mobile and Vehicle Systems

Xiaochen Li, Sicong Liu, Bin Guo, Yu Ouyang, Fengmin Wu, Yuan Xu, Zhiwen Yu

Abstract

GB-scale large apps like on-device LLMs and rich media editors are becoming the next-generation trend, but their heavy memory and I/O demands, especially during multitasking, cause devices to reclaim or kill processes, turning warm apps into cold launches. The challenge lies not in storing them, but in fast, accurate launching. For users, 1s is the usability cliff, yet our measurements show 86.6\% of GB-scale cold launches exceed it. Also, Android Vitals flags only $\geq$ 5s as slow, exposing a large satisfaction gap. Existing optimizations are designed in isolation and conflict. For example, preloading reduces I/O stalls but consumes scarce memory and is undone by reclamation, while reclamation and killing free memory but sacrifice background survivability, leading to repeated cold relaunches. Our key insight is that, although multitasking makes runtime behavior complex, each app's file access pattern remains predictable. The challenge lies in exploiting this predictability, i.e., preloading without exhausting memory, reclaiming without undoing gains, and killing selectively to preserve background survivability. We introduce AppFlow, a prediction-based system-wide scheduler that integrates a Selective File Preloader, an Adaptive Memory Reclaimer, and a Context-Aware Process Killer. Implemented across the Android framework and Linux kernel without app changes, AppFlow cuts GB-scale cold-launch latency by 66.5\% (e.g., 2s$\rightarrow$690ms) and sustains 95\% of launches within 1s over a 100-day test, significantly improving responsiveness and multitasking experience.

AppFlow: Memory Scheduling for Cold Launch of Large Apps on Mobile and Vehicle Systems

Abstract

GB-scale large apps like on-device LLMs and rich media editors are becoming the next-generation trend, but their heavy memory and I/O demands, especially during multitasking, cause devices to reclaim or kill processes, turning warm apps into cold launches. The challenge lies not in storing them, but in fast, accurate launching. For users, 1s is the usability cliff, yet our measurements show 86.6\% of GB-scale cold launches exceed it. Also, Android Vitals flags only 5s as slow, exposing a large satisfaction gap. Existing optimizations are designed in isolation and conflict. For example, preloading reduces I/O stalls but consumes scarce memory and is undone by reclamation, while reclamation and killing free memory but sacrifice background survivability, leading to repeated cold relaunches. Our key insight is that, although multitasking makes runtime behavior complex, each app's file access pattern remains predictable. The challenge lies in exploiting this predictability, i.e., preloading without exhausting memory, reclaiming without undoing gains, and killing selectively to preserve background survivability. We introduce AppFlow, a prediction-based system-wide scheduler that integrates a Selective File Preloader, an Adaptive Memory Reclaimer, and a Context-Aware Process Killer. Implemented across the Android framework and Linux kernel without app changes, AppFlow cuts GB-scale cold-launch latency by 66.5\% (e.g., 2s690ms) and sustains 95\% of launches within 1s over a 100-day test, significantly improving responsiveness and multitasking experience.
Paper Structure (39 sections, 8 equations, 21 figures, 1 table, 1 algorithm)

This paper contains 39 sections, 8 equations, 21 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Primer of App Launch Processes
  4. The Trend of GB‑Scale Large Mobile Apps
  5. GB-scale Large App Cold-Launch Problems
  6. Mobile Memory Management.
  7. Observation
  8. Idle I/O and Hot-set File
  9. Page-varying Reclamation Cost
  10. Temporal Killing Impact
  11. Preload–Reclaim Interference
  12. Overview
  13. Problem Formulation
  14. Solution Overview
  15. System Design
  16. ...and 24 more sections

Figures (21)

  • Figure 1: Over the past few years, the share of GB-scale apps in the Google Play top 10 has grown markedly, alongside a rapid surge in DRAM prices.
  • Figure 2: Illustration of GB-scale app launch progress.
  • Figure 3: Illustration of launch latency across cold, warm, and hot modes for 15 popular GB-scale apps.
  • Figure 4: Impact of memory availability on app launch time: comparing cold‑launch latencies with 10 background apps (<1GB free memory) vs. no background apps (>2GB free).
  • Figure 5: During GB-scale large app launch, small-file($\leq$128 KB) accesses outnumber large-file($>$128 KB) accesses by 4.7$\times$ yet occupy just 2.2% of their memory.
  • ...and 16 more figures