International Journal of Network Management

Cover image for Vol. 27 Issue 1

Edited By: James Won-Ki Hong

Impact Factor: 0.681

ISI Journal Citation Reports © Ranking: 2015: 65/82 (Telecommunications); 111/144 (Computer Science Information Systems)

Online ISSN: 1099-1190

Recently Published Issues

See all

International Journal of Network Management

The International Journal of Network Management is the forum for researchers, developers, and practitioners in network management to present their work to an international audience. The journal is dedicated to the dissemination of information, which will enable improved management, operation, and maintenance of computer networks and communication systems.

Click here to read more about this journal.

Recently Published Articles

  1. You have free access to this content
    An accurate traffic classification model based on support vector machines

    Jie Cao, Zhiyi Fang, Guannan Qu, Hongyu Sun and Dan Zhang

    Version of Record online: 9 JAN 2017 | DOI: 10.1002/nem.1962

    Thumbnail image of graphical abstract

    The traffic classification model is deducted from the scaling dataset and uses principal component analysis (PCA) to extract data features and verify its relevant traffic features obtained from PCA. The optimal working parameters of kernel function are derived automatically from improved particle swarm optimization (PSO) algorithm. Then, using cross-validation method to train and generate the SPP-SVM classifier model, classing the test set, and getting the classification results based on the SPP-SVM classifier model.

  2. You have free access to this content
    Header signature maintenance for Internet traffic identification

    Sung-Ho Yoon, Jun-Sang Park, Baraka D. Sija, Mi-Jung Choi and Myung-Sup Kim

    Version of Record online: 16 DEC 2016 | DOI: 10.1002/nem.1959

    Thumbnail image of graphical abstract

    In this article, we propose an efficient method for header signature maintenance. Our approach automatically constructs header signatures for traffic identification and only retains the most significant signatures in the signature repository to save memory space and to improve matching speed. For the signature maintenance, we define a new metric, the so-called signature weight, that reflects its potential ability to identify traffic. Signature weight is periodically calculated and updated to adapt to the changes of network environment.

  3. A dynamic pricing algorithm for a network of virtual resources

    Bram Naudts, Mario Flores, Rashid Mijumbi, Sofie Verbrugge, Joan Serrat and Didier Colle

    Version of Record online: 12 DEC 2016 | DOI: 10.1002/nem.1960

    Thumbnail image of graphical abstract

    This article proposes a dynamic pricing algorithm for networks of virtual resources. The algorithm determines at which utilization level it is rewarding to charge a higher price for a particular resource and the alternative price that should be charged. Applying this algorithm results in higher total revenue compared to the traditional static pricing approach.

  4. You have free access to this content
    A tree-based algorithm for virtual infrastructure allocation with joint virtual machine and network requirements

    Ramon de Oliveira and Guilherme Piegas Koslovski

    Version of Record online: 29 NOV 2016 | DOI: 10.1002/nem.1958

    Thumbnail image of graphical abstract

    Aiming for virtual infrastructures allocation with joint virtual machine and network requirements atop cloud datacenters,we propose a heuristic having as premise that both graphs (virtual and physical) are represented in the form of trees (the graphs are connected and acyclic). Following this premise, strategies to reduce search space by grouping subtrees information are applied, which consequently accelerates the allocation process. Initially, the mechanism, termed VITreeM, converts both physical and virtual graphs in trees. In a second moment, the search space reduction algorithm resumes capacity information of subtrees, consequently removing the need for a deep search as a capacity summary is available at each tree element. Finally, VITreeM performs the allocation by comparing the abstracted representation of subtrees.

  5. Fast failure detection and recovery in SDN with stateful data plane

    Carmelo Cascone, Davide Sanvito, Luca Pollini, Antonio Capone and Brunilde Sansò

    Version of Record online: 22 NOV 2016 | DOI: 10.1002/nem.1957

    Thumbnail image of graphical abstract

    We present SPIDER, a design for a packet forwarding pipeline to perform sub-millisecond failure detection and rerouting directly in the switch fast-path, ie, without requiring the intervention of the slow control plane. SPIDER is based on recent advances in stateful data plane abstractions for software-defined networking, such as OpenState or P4. We present prototype implementations and numerical results on switches' memory impact and performances in recovery latency and packet loss.