The rapid increase in users of application (App) for smart device during disasters and emergencies necessitates realistic and large-scale simulations. Existing methods using physical devices or emulators face challenges in hardware dependency and overhead, making the simultaneous execution of millions of App difficult. This study proposes a method for constructing “vApp,” headless App that can be executed as back-end processes independent of GUI. By leveraging design patterns of GUI application, the front-end is rebuilt while retaining the logic for processing user inputs. Collaborating with a multi-agent simulator (Agent) allows appropriate simulation of user operations and environmental information, enabling the construction of a large-scale automated execution environment. Evaluation results demonstrate that the proposed method outperforms conventional methods in terms of computational resource efficiency, time reduction, execution cost savings, and scalability.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Structure for Smart Device Applications for Large Scale Simulation

  • Soma Nakagawa,
  • Satoshi Uda,
  • Yoichi Shinoda

摘要

The rapid increase in users of application (App) for smart device during disasters and emergencies necessitates realistic and large-scale simulations. Existing methods using physical devices or emulators face challenges in hardware dependency and overhead, making the simultaneous execution of millions of App difficult. This study proposes a method for constructing “vApp,” headless App that can be executed as back-end processes independent of GUI. By leveraging design patterns of GUI application, the front-end is rebuilt while retaining the logic for processing user inputs. Collaborating with a multi-agent simulator (Agent) allows appropriate simulation of user operations and environmental information, enabling the construction of a large-scale automated execution environment. Evaluation results demonstrate that the proposed method outperforms conventional methods in terms of computational resource efficiency, time reduction, execution cost savings, and scalability.