ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

High performance computing in Windows Azure cloud

Dejan Ambruš (2013) High performance computing in Windows Azure cloud. EngD thesis.

Download (1637Kb)


    High performance, security, availability, scalability, flexibility and lower costs of maintenance have essentially contributed to the growing popularity of cloud computing in all spheres of life, especially in business. In fact cloud computing offers even more than this. With usage of virtual computing clusters a runtime environment for high performance computing can be efficiently implemented also in a cloud. There are many advantages but also some disadvantages of cloud computing, some of them are addressed in our thesis, mainly regarding to the Microsoft's public cloud computing platform Windows Azure. We describe the underlying infrastructure and software configuration as well as its functions and server usage charges. In the second part we describe how to set up a Windows Azure account and configure it for high-performance computing and also how to run many virtual machines simultaneously using HPC Cluster Manager. Dynamic resource allocation, improved flexibility, broad network access, performance enhancements as well as business benefits convinced us there are many benefits of cloud computing comparing to the on-site computing resources.

    Item Type: Thesis (EngD thesis)
    Keywords: High performance computing, Windows Azure, cloud, virtual cluster
    Number of Pages: 58
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    izr. prof. dr. Uroš Lotrič270Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=9685332)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 1963
    Date Deposited: 15 Jan 2013 13:04
    Last Modified: 28 Feb 2013 13:34
    URI: http://eprints.fri.uni-lj.si/id/eprint/1963

    Actions (login required)

    View Item