North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

SIGCOMM Workshop on Network Troubleshooting

  • From: jon bennett
  • Date: Tue Feb 24 20:41:52 2004


(Recent messages on the list suggest this announcement to be quite timely)

We hope that this workshop will be of interest to NANOG members and that many of you will participate and/or submit material to it.

Jon Bennett and Mark Allman ( [email protected] )

http://www.acm.org/sigcomm/sigcomm2004/netts.html

-----------------------------------------------------------------------
To be held in conjunction with SIGCOMM '04

Network Troubleshooting:
Research Theory and Operations Practice Meet Malfunctioning Reality

Call For Papers

Network monitoring and measurement has received a great deal of attention in the research community recently. While some research to-date has been focused on finding problems, failures and anomalies in networks this workshop endeavors to focus specifically on such topics. The workshop seeks papers exploring several themes:

DETECTION: Mechanisms and techniques for detecting failures, imminent failures and other anomalies in real time. The focus of this workshop is research that can be used operationally to help the network in the short-term. Techniques that require heavyweight off-line analysis to find problems provide the community with an understanding of and an insight into the dynamics and potential long-term solutions for network issues, but are not the main focus of this workshop.

CORRECTION: While detecting problems (or imminent problems) and alerting network operators is a good first step, techniques for automatically mitigating problems as they occur are also sought.

COORDINATION: Detecting and solving problems in a multi-provider environment inevitably involves communicating between distinct autonomous entities. Mechanisms and facilities to streamline and automate such communication are sought.

EXPERIENCE: Insight from network operators into network problems they cannot easily detect (or, detect far too late) and tools that would make network management much easier. Input from network operators on non-obvious or non-technical considerations which impact technical solutions are also sought.

This workshop invites two kinds of submissions:

Original papers on any area of network measurement, monitoring or management specifically directed towards one or more of the above themes.

Poster presentation proposals. While posters on any of the above themes will be accepted, posters on operational experience are highly sought.


Note: For this workshop, "networks" includes both physical networks and virtual networks (CDNs, overlays, etc.).

Some of the specific problems of interest, include:
Protocol failures - link, routing, management
Detecting mis-configuration of network elements
Partial hardware failures - intermittent, unreported
Traffic engineering for overload control
Security - DDoS attacks, detecting compromised network elements, intrusion detection (especially for non-edge networks since a large body of work already tackles the problems at the network edge)

Submissions:


Submissions ranging from presentations of specific research to position papers are welcome. Papers presenting interesting and novel ideas at an early stage of development are preferred over completed journal-style results. Selected papers will be forward-looking, with impact and implications for both operational networks and ongoing or future research.

Original papers should be 3-6 standard SIGCOMM formatted pages (with the expectation that position papers will be shorter and research papers longer).

Poster proposals should be sent in the form of 1-page abstracts.

The submission process and specific guidelines will be posted at a later date.

Important Dates:

Submission registration: April 8, 2004
Submission deadline: April 15, 2004
Notification deadline: May 15, 2004
Camera ready deadline: June 15, 2004


Workshop Co-Chairs ([email protected])

Jon C.R. Bennett, Harvard University
Mark Allman, ICIR

Program Committee:

TBD