Workshop on the Analysis of System Logs - Oct 14 - Call for Participation

        Workshop on the Analysis of System Logs (WASL) 2009
                http://www.systemloganalysis.com
                    Call for Participation

                 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
                       October 14, 2009
                         Big Sky, MT
                           (at SOSP)
                 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

-------------------------------------------------------------------------=
-

System logs contain a wide variety of information about system status=20
and health,
including events from various applications, daemons and drivers, as well=20
as sampled
information such as resource utilization statistics. As such, these logs=20
represent a
rich source of information for the analysis and diagnosis of system=20
problems and
prediction of future system events. However, their lack of organization=20
and the general
lack of semantic consistency between information from various software=20
and hardware
vendors means that most of this information content is wasted. Indeed,=20
today's
most popular log analysis technique is to use regular expressions to=20
either detect
events of interest or to filter the log so that a human operator can=20
examine it manually.
Clearly, this captures only a fraction of the information available in=20
these logs and
does not scale to the large systems common in business and=20
supercomputing environments.
This workshop will focus on novel techniques for extracting=20
operationally useful
information from existing logs and methods to improve the information=20
content of future
logs.

Workshop Program

Session 1: Log Analysis Tools
    - "Extracting Message Types from BlueGene/L's Logs", A. Makanju, A.=20
Zincir-Heywood, and E. Milios =20
    - "Incremental Learning of System Log Formats", K. Zhu, K. Fisher,=20
and D. Walker =20
    - "Visual and Algorithmic Tooling for System Trace Analysis: A Case=20
Study", W. De Pauw and S. Heisig =20

Session 2: Analyzing System Logs
    - "Mining Dependency in Distributed Systems through Unstructured=20
Logs Analysis", J. Lou, Q. Fu, Y. Wang, and J. Li =20
    - "A Bayesian Network Approach to Modeling IT Service Availability=20
using System Logs", R. Zhang, E. Cope, L. Huesler, and F. Cheng =20
    - "Endpoint Identification Using System Logs", S. Melvin =20

Session 3: Group Discussion on Current State of the Art
    - Tips and tricks in current use.
    - Gaps and challenges in current techniques.
    - Vision and steps for the future.
=20
Session 4: Panel on Future Research Agenda
    - What are the most difficult problems with logging, in the real worl=
d?
    - How to make academia-industry interactions more productive?
    - How to extract meaningful information from logs?
    - How to improve system management?
=20

Workshop Chair:
    Greg Bronevetsky (Lawrence Livermore National Laboratory)
    [email protected]

Program Committee:
    Jon Stearley, Sandia National Laboratory
    Bianca Schroeder, University of Toronto
    S=E9bastien Tricaud, INL
    Sapan Bhatia, Princeton University
    Risto Vaarandi, CCD CoE
    Jim Jansen, Penn State University
    Wei Xu, University of California, Berkeley
    Anton Chuvakin, Qualys
    Kara Nance, University of Alaska, Fairbanks
    Raffael Marty, PixlCloud