• DocumentCode
    3746666
  • Title

    Tutorial on a modeling and simulation cloud service

  • Author

    Daniel Zehe;Wentong Cai;Alois Knoll;Heiko Aydt

  • Author_Institution
    TUM CREATE, 1 CREATE WAY, Singapore 138602
  • fYear
    2015
  • Firstpage
    103
  • Lastpage
    114
  • Abstract
    For large-scale urban system simulations the computing power of traditional workstations is not sufficient. The move to High Performance Computing clusters is a viable solution. Users of such simulations are domain experts with little knowledge in computer science and optimization of such simulations. The access to HPC resources is also not available. Vendors have not sufficiently addressed this. This leads to the conclusion of moving the computational demand to the cloud, where the on-demand culture for resources has been expanding. In this tutorial we will present an approach of how to work with an entirely cloud-based solution for modeling and simulation, with an exemplary implementation of an urban traffic simulation cloud service. Since the computational offload from the workstation to a remote computing entity also allows the use of novel user interfaces (design and devices), through the use of RESTful interfaces, use-case applicable interfaces for simulations can also be created.
  • Keywords
    "Computational modeling","Cloud computing","Data models","Hardware","Workstations","Runtime"
  • Publisher
    ieee
  • Conference_Titel
    Winter Simulation Conference (WSC), 2015
  • Electronic_ISBN
    1558-4305
  • Type

    conf

  • DOI
    10.1109/WSC.2015.7408156
  • Filename
    7408156