CFP: Deadline Extended: SLAML'10

       Workshop on Managing Systems via Log Analysis and Machine=20
                  Learning Techniques (SLAML '10)

             =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=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
                          October 2-3, 2010
                        Vancouver, BC, Canada
                             (at OSDI)
               http://www.usenix.org/events/slaml10/cfp/
             =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=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

   =20
              ********   DEADLINE EXTENDED  ***********
   Now accepting full both 8-page full papers and 3-page position papers!

           FULL PAPER SUBMISSION: Sunday, July 11, 2010, 11:59 p.m. PDT
           AUTHOR NOTIFICATION: Friday, August 20, 2010
           FINAL PAPERS DUE: Thursday, September 16, 2010

    SLAML '10 combines the Workshop on the Analysis of System Logs (WASL) a=
nd the Workshop on Tackling Computer Systems Problems with Machine Learning=
 Techniques (SysML). We welcome contributions related to either of these im=
portant and related topics.

    Modern large-scale systems are challenging to manage. Fortunately, as t=
hese systems generate massive amounts of performance and diagnostic data, t=
here is an opportunity to make system administration and development simple=
r via automated techniques to extract actionable information from the data.=
 This workshop addresses this problem in two thrusts: (i) the analysis of r=
aw system data logs and (ii) the application of machine learning to systems=
 problems. We expect the large overlap in these topics to promote a rich in=
terchange of ideas between the areas.

Log Analysis:=20
    It is well known that raw system logs are an abundant source of informa=
tion for the analysis and diagnosis of system problems and prediction of fu=
ture system events. However, a lack of organization and semantic consistenc=
y between system data from various software and hardware vendors means that=
 most of this information content is wasted. Current approaches to extracti=
ng information from the raw system data capture only a fraction of the info=
rmation available and do not scale to the large systems common in business =
and supercomputing environments. It is thus a significant research challeng=
e to determine how to better process and combine information from these dat=
a sources.

Machine Learning:=20
    The large scale of available data requires automated and machine-assist=
ed analysis. Statistical machine learning techniques have recently shown gr=
eat promise in meeting the challenges of scale and complexity in datacenter=
-scale and Internet-scale computing systems. However, applying these techni=
ques to real systems scenarios requires careful analysis and engineering of=
 the techniques to fit them to specific scenarios; there is sometimes also =
the opportunity to develop new algorithms specific to systems scenarios. Th=
is workshop thrust thus also presents a substantial research area: the expl=
oration of new approaches to using machine learning to help us understand, =
measure, and diagnose complex systems.

Topics include but are not limited to:
    o Reports on publicly available sources of sample system logs
    o Prediction of malfunction or misuse based on system data
    o Statistical analysis of system logs
    o Applications of Natural-Language Processing (NLP) to system data
    o Techniques for system log analysis, comparison, standardization, comp=
ression, anonymization, and visualization
    o Applications of log analysis to system administration problems
    o Use of machine learning techniques to address reliability, performanc=
e, power management, security, fault diagnosis, scheduling, or manageabilit=
y issues
    o Challenges of scale in applying machine learning to large systems
    o Integration of machine learning into real-world systems and processes
    o Evaluating the quality of learned models, including assessing the con=
fidence/reliability of models and comparisons between different methods


    SLAML '10 will be a 1.5-day workshop held immediately preceding the 9th=
 USENIX Symposium on Operating Systems Design and Implementation (OSDI '10)=
, which will take place October 4-6, 2010. SLAML'10 will begin on the after=
noon of October 2, 2010, and run through October 3, 2010.


Workshop Organizers:=20
    Greg Bronevetsky, Lawrence Livermore National Laboratory
    Kathryn Mohror, Lawrence Livermore National Laboratory
    Alice Zheng, Microsoft Research

Submission Instructions:
    Interested speakers should submit their full papers by June 13, 2010, v=
ia the Web submission form, which will be available here soon. All papers w=
ill be subject to peer review under conference standards. Authors may choos=
e to submit a paper anonymously or with author names visible to reviewers. =
Experience reports and papers on work in progress are welcome as long as th=
ere is a clear contribution. Paper submissions may be accepted as papers fo=
r the regular workshop program or as posters for the poster session. Submis=
sions must be in PDF format and must be no longer than eight 8.5" x 11" pag=
es, including figures and tables, but not including references, formatted i=
n two columns, using 10 point type on 12 point (single-spaced) leading, wit=
h the text block being no more than 6.5" wide by 9" deep.
    All papers will be available online to registered attendees before the =
workshop. If your accepted paper should not be published prior to the event=
, please notify [email protected]. The papers will be available online =
to everyone beginning on the first day of the workshop, October 2, 2010.
    Papers accompanied by nondisclosure agreement forms will not be conside=
red. Accepted submissions will be treated as confidential prior to publicat=
ion on the USENIX SLAML '10 Web site; rejected submissions will be permanen=
tly treated as confidential.
    Simultaneous submission of the same work to multiple venues, submission=
 of previously published work, or plagiarism constitutes dishonesty or frau=
d. USENIX, like other scientific and technical conferences and journals, pr=
ohibits these practices and may take action against authors who have commit=
ted them. See the USENIX Conference Submissions Policy for details.

   Questions? Contact your program co-chairs, [email protected], or =
the USENIX office, [email protected].