Event Abstract

Back to Event

Simulation Platform: Model simulation on the cloud

Shiro Usui1, Tadashi Yamazaki1, Hidetoshi Ikeno2*, Okumura Yoshihiro3, Sato Shunji4, Kamiyama Yoshimi5, Hirata Yutaka6, Inagaki Keiichiro1, Ishihara Akito7, Kannon Takayuki1, Kamiji Nilton1 and Akazawa Fumihiko3
  • 1 RIKEN BSI, Japan
  • 2 University of Hyogo, School of Human Science and Environement, Japan
  • 3 NIJC, RIKEN, Japan
  • 4 University of Electro-Communications, Graduate School of Informatis Systems, Japan
  • 5 Aichi Prefectural University, School of Information Science and Technology, Japan
  • 6 Chubu University, Faculty of Engineering, Japan
  • 7 Chukyo University, School of Information Science and Technology, Japan

Various kinds of data analysis methods and computational models have been proposed and utilized in neuroscience fields. In this regard, theoretical approach is indispensable for further advance in neuroscience. Recently, a number of software packages for data analysis and scripts/programs for computational models have been provided on the Internet, and many of them are open-access for researchers to validate and reuse. Neuroinformatics Japan Center (NIJC) has been supporting these activities and hosting neuroinformatics database sites called platforms. The platforms contain various research resources, such as experimental data, data analysis tools and computational model scripts/programs as well as research papers to share the knowledge and technique for accelerating computational neuroscience researches. Although a number of resources are available on the platforms and other databases, preparations to review experimental data and execute model simulation on ones own computer still remain burdensome. If a model script is constructed on a neural simulator such as NEURON or GENESIS, one has to install these simulators in our computers even just for a trial use. If a model is written in programming language such as C, C++ or Python, one has to prepare the programming environment including compiler like gcc, g++ and other tools and libraries. Furthermore, if a model is provided in a commercial product, such as MATLAB or Mathematica, one may have to purchase even just on trial. Therefore, we propose a novel framework to avoid these difficulties and provide user-friendly environment to execute model simulation and data analysis programs registered in the platforms without any preparation on a user’s computer. Here, we present our on-going project on launching a web service called Simulation Platform (SimPF) as one of the Japan-node platforms (http://www.neuroinf.jp). SimPF provides two types of virtual computing environment for neuroscientists via web, one is virtual machine (VM) and another is grid computing. VM environment is realized by VirtualBox (http://www.virtualbox.org/) running on CentOS. A variety of neural simulators, application and developmental tools, including NEURON, GENESIS, ImageJ, C, C++ and MATLAB, are installed in VM. By using VM, users can review the experimental data with less effort and run the computational model programs that are registered on the platform and database such as ModelDB. It is no need to install any application software or compiler on own computer. That is, when a user clicks the start button of VM, appropriate model program with control script for automatic execution (auto-run script) is sent to the SimPF. Once these files are uploaded, SimPF assigns a VM for the user from SimPF clouds and connects the VM automatically to the user's browser via VNC (Virtual Network Computing) protocol. The desktop screen of the VM appears on the browser, and then the auto-run script starts model simulation automatically. It is true that computational neuroscientists want to have high performance computing environment, such as grid computing, in their laboratories. Usually, each laboratory has several computers, but it is still needed to have technical know-how for constructing own cloud computing system. SimPF also provides a grid computing environment and several model programs for it. The SimPF grid environment is based on open source software such as Globus toolkit and Ganglia. The SimPF, not only lets users be free from preparing the environment to carry out simulation, but also allows simulation for reviewing and analyzing existing models, as well as for reviewing newly presented model submitted for publication. We expect that the SimPF will further support education and research in neuroscience fields.

Conference: Neuroinformatics 2010 , Kobe, Japan, 30 Aug - 1 Sep, 2010.

Presentation Type: Oral Presentation

Topic: Infrastructural and portal services

Citation: Usui S, Yamazaki T, Ikeno H, Yoshihiro O, Shunji S, Yoshimi K, Yutaka H, Keiichiro I, Akito I, Takayuki K, Nilton K and Fumihiko A (2010). Simulation Platform: Model simulation on the cloud. Front. Neurosci. Conference Abstract: Neuroinformatics 2010 . doi: 10.3389/conf.fnins.2010.13.00100

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 14 Jun 2010; Published Online: 14 Jun 2010.

* Correspondence: Hidetoshi Ikeno, University of Hyogo, School of Human Science and Environement, Himeji, Japan, ikeno@shse.u-hyogo.ac.jp