publications:
- title: "Performance Analysis of Distributed Server Systems"
  author:
  - name: "Greg Franks"
    link: "http://www.sce.carleton.ca/faculty/franks"
  - name: "Shikharesh Majumdar"
    link: "http://www.sce.carleton.ca/faculty/majumdar"
  - name: "Neilson, John"
    link: "https://researchr.org/alias/neilson%2C-john"
  - name: "Dorina Petriu"
    link: "http://www.sce.carleton.ca/faculty/petriu"
  - name: "Jerome Rolia"
    link: "http://www.hpl.hp.com/people/jerry_rolia/"
  - name: "Murray Woodside"
    link: "http://www.sce.carleton.ca/faculty/woodside"
  year: "1996"
  month: "oct"
  abstract: "It is generally accepted that performance characteristics, such as response time and throughput, are an integral part of the factors defining the quality of software products. The relationship between quality and system responsiveness is especially strong in the case of distributed application using various kind of software servers (name servers, application servers, database servers, etc.) In order to meet the performance requirements of such systems, the developers should be able to assess and understand the effect of various design decisions on system performance at an early stage, when changes can be made easily and effectively. Performance analysis should then continue throughout the whole life cycle, becoming one of the means of assuring the quality of software products. For this to become a practical reality, we need appropriate modeling techniques.$\\backslash$$\\backslash$ This paper presents a new performance model named \\{$\\backslash$em Layered Queueing Networks\\} (LQN) for systems with distributed software servers. In such systems the servers are are frequently layered, so that lower level delays are included in the higher layer service time, which limits the system capacity. LQN represents these effects in a compact and understandable way, and provide analytical and simulation tools. The paper explains the model, and more importantly, shows how can be applied to identify performance trouble spots in a system and devise effective corrective measures."
  tags:
  - "meta-model"
  - "modeling"
  - "software product quality"
  - "analysis"
  - "database"
  - "Meta-Environment"
  - "design"
  researchr: "https://researchr.org/publication/perf%3Afranks-96"
  cites: 0
  citedby: 0
  pages: "15-26"
  booktitle: "The Sixth International Conference on Software Quality (6ICSQ)"
  organization: "American Society for Quality Control (ASQC)"
  kind: "inproceedings"
  key: "perf:franks-96"