• DocumentCode
    607985
  • Title

    Grid on Cloud

  • Author

    Kibe, S. ; Yamagiwa, M. ; Uehara, M.

  • Author_Institution
    Dept. of Open Inf. Syst., Toyo Univ., Kawagoe, Japan
  • fYear
    2013
  • fDate
    25-28 March 2013
  • Firstpage
    213
  • Lastpage
    218
  • Abstract
    Recently, dramatic developments have taken place in cloud computing, requiring IT specialists to acquire new cloud skills. Accordingly, universities will need to change their curriculums and educational environments in order to teach advanced information and communication technology to students. Educational organizations such as universities need to provide an educational cloud for use by students. We are constructing a private educational cloud based on the Ubuntu Enterprise Cloud. The most important feature of our cloud is super-saturation, which is defined as allocating more logical resources than the available physical resources. A supersaturated cloud can execute more than 10 times the number of instances than a conventional cloud. In order to maximize the performance of a supersaturated cloud, tasks must continuously be provided to the cloud. However, the demand on grids has been reduced because of energy saving. Thus, the grid and cloud are complementary technologies in that the cloud consumes tasks produced by the grid. In this paper, we propose a method for constructing a Grid on Cloud to solve this issue. We compare two methods, the single large instance strategy (SLIS) and multiple small instance strategy. We conclude that the SLIS based instance grid method is the best if the system load is shared by all instances.
  • Keywords
    cloud computing; computer aided instruction; computer science education; educational courses; grid computing; resource allocation; software performance evaluation; teaching; SLIS based instance grid method; Ubuntu Enterprise Cloud; advanced information and communication technology; cloud computing; cloud skills; curriculums; educational environments; energy saving; logical resource allocation; multiple small instance strategy; physical resources; private educational cloud; single large instance strategy; supersaturated cloud performance; system load; teaching; Business; Cloud computing; Clouds; Educational institutions; Memory management; Servers; cloud computing; educational cloud; grid;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Advanced Information Networking and Applications (AINA), 2013 IEEE 27th International Conference on
  • Conference_Location
    Barcelona
  • ISSN
    1550-445X
  • Print_ISBN
    978-1-4673-5550-6
  • Electronic_ISBN
    1550-445X
  • Type

    conf

  • DOI
    10.1109/AINA.2013.71
  • Filename
    6531758