• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 10
  • 2
  • Tagged with
  • 12
  • 12
  • 12
  • 12
  • 7
  • 6
  • 6
  • 6
  • 6
  • 6
  • 6
  • 6
  • 6
  • 6
  • 5
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
11

Differentiated dependability of the 3LIHON concept with main focus on internal node architecture

Fernandez-Fernandez, Alvaro January 2012 (has links)
Hybrid optical networks are considered as the most promising architectures for future optical networks, in order to achieve both better cost and performance. Dependability is a topic of increasing interest in order to provide the Quality of Service (QoS) expected for future optical networks. Recently, the 3-Level Integrated Hybrid Optical Network (3-LIHON) has been proposed as a new concept for future optical core networks, including different transport technologies to support a wide range of services. Due to the wide range of future telecommunication services foreseen for next generation optical backbone networks, the 3-LIHON architecture must be provided with differentiated dependability, in order to offer the availability demanded by the most critical services, as well as to utilize network resources efficiently. By focusing in a single 3-LIHON node, this thesis employs component redundancy to present a differentiated survivability model suitable for the 3-LIHON architecture. Several protection mechanisms for two of the three types of traffic supported by 3-LIHON are proposed in this research work. These mechanisms are analyzed by means of different dependability models, i.e. reliability block diagrams and Markov models; and a numerical evaluation of its unavailability is presented. In addition, simulation results measuring the impact on performance of some mechanisms are also presented. Then, the protection mechanisms are compared and evaluated, taking into account not only its unavailability and performance, but also other considerations such as cost and complexity. Finally, based on this evaluation, the best suited mechanisms are selected.
12

Design and Implementation of an Efficient, Reliable and Safe Work-Package Database System at CERN

Riksem, Hans-Even Ramsevik January 2010 (has links)
The Activity Coordination Tool (ACT) is a web application designed to automate theplanning and coordination of work packages. In the CMS experiment at CERN it isimportant that work packages in the underground facilities are properly planned inorder to not jeopardize time schedules, equipments, budgets and safety. The subjectof this thesis is the development of the database schema used by the ACT application.The schema has been developed from scratch in order to best fit the needs of CMS andto cover all aspects of the planning and coordination process not found in other CERNdatabases.Models and diagrams of the database schema are based on a step by step descriptionof the work package process. This step by step description was used to organize thedata, to make the data and data relationships consistency, and to make the databasestructure flexible for extensions in the future. These models would eventually be usedto implement the schema in the CMS online database.Another database at CERN has many similarities with the ACT database. These twodatabases will eventually merge due to common interest, and the ACT database schemaneeded therefore to resemble some tables in this database in order to facilitate themerging. Some problems with network traversal and security needs to be solved beforethese two databases can be merged.The architecture and usage of the database schema can affect the database performancebut it doesn’t contribute to the reliability of the databases system. The hardware andsoftware components making up the database system itself are usually the main contributorsto this. The CMS online network work hard to keep the performance and reliabilityof their database system as good as possible. Everything from disks to networkconnections is redundant. In addition to component redundancy are features providedby Oracle used to improve performance. The amount of server redundancy does howeverseem a bit exaggerated, and the performance could be slightly improved if otherOracle features were used.

Page generated in 0.131 seconds