• DocumentCode
    153244
  • Title

    Modeling and Analysis of Availability for SaaS Multi-tenant Architecture

  • Author

    Wenbo Su ; Chuang Lin ; Kun Meng ; Qu Liu

  • Author_Institution
    Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing, China
  • fYear
    2014
  • fDate
    7-11 April 2014
  • Firstpage
    365
  • Lastpage
    369
  • Abstract
    Software as a Service(SaaS) is increasingly becoming an important application development and delivery model. Among different implementation architectures, multi-tenant architecture(MTA) has received most attentions. Not only because it is easier to be maintained, but also increases the resource utilization. However, sharing the same instance will have impact on the availability of each tenant. Availability is a very critical attribute of tenants´ SLA, but modeling and analysis of availability for MTA SaaS is challenging. In this paper, we firstly discuss the differences of availability between one-tenant and multi-tenant from tenant´s view. Then we analyze the repair and failure rate based on the MTA feature and give the function expression with parameters of the number of tenant´s users. And a Markov model is built for availability analysis. We solve the model and get the computation expression of the availability in a case study. Then a numerical result is obtained under the specific conditions and finally we have a simply discussion on it.
  • Keywords
    Markov processes; cloud computing; resource allocation; Markov model; SaaS multitenant architecture; application development; availability analysis; delivery model; resource utilization; software as a service; Analytical models; Availability; Computational modeling; Computer architecture; Databases; Software as a service; Software reliability; SaaS; multi-tenant; availability; modeling; Markov process;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Service Oriented System Engineering (SOSE), 2014 IEEE 8th International Symposium on
  • Conference_Location
    Oxford
  • Type

    conf

  • DOI
    10.1109/SOSE.2014.65
  • Filename
    6830931