Managing networks through context: Graph visualization and exploration

Qi Liao, Andrew Blaich, Dirk Vanbruggen, Aaron Striegel

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

With the increasing prevalence of multi-user environments in distributed systems, it has become an increasingly challenging task to precisely identify who is doing what on an enterprise network. Current management systems that rely on inference for user identity and application are not capable of accurately reporting and managing a large-scale network due to the coarseness of the collected data or scaling of the collection mechanism. We propose a system that focuses data collection in the form of local context, i.e. the precise user and application associated with a network connection. Through the use of dynamic correlation and novel graph modeling, we developed a visualization tool called ENAVis (the work appeared in earlier form in [1] and received USENIX best paper award). (Enterprise Network Activities Visualization). ENAVis aids a real-world administrator in allowing them to more efficiently manage and gain insight about the connectivity between hosts, users, applications and data access offering significant streamlining of the management process.

Original languageEnglish
Pages (from-to)2809-2824
Number of pages16
JournalComputer Networks
Volume54
Issue number16
DOIs
StatePublished - Nov 15 2010

Keywords

  • Context
  • Enterprise network management
  • Forensics
  • Graphs
  • Interactive exploration
  • Security
  • Visual mining
  • Visualization

Fingerprint

Dive into the research topics of 'Managing networks through context: Graph visualization and exploration'. Together they form a unique fingerprint.

Cite this