Check out the new USENIX Web site.

Second Workshop on Tackling Computer Systems Problems
with Machine Learning Techniques (SysML07)

April 10, 2007, Cambridge, MA
Co-located with


Program Co-chairs

Program Committee

Steering Committee

  • Emre Kiciman, Microsoft Research
  • Moises Goldszmidt, Microsoft Research


The online proceedings are now available below.
Note: Until a year after the workshop, the full refereed papers are available only to USENIX members. When accessing restricted papers, remember to use your member number or login name as your userid. Not a USENIX member? Join today!

8:30 a.m.–9:00 a.m.  

Welcome and Introduction to the Workshop

9:00 a.m.–10:45 a.m.  

Session I

Machine Learning for Efficient Neighbor Selection in Unstructured P2P Networks
Robert Beverly, MIT CSAIL; Mike Afergan, Akamai/MIT
Read the paper in PDF | HTML

Gray's Anatomy: Dissecting Scanning Activities Using IP Gray Space Analysis
Yu Jin, György Simon, Kuai Xu, Zhi-Li Zhang, and Vipin Kumar, University of Minnesota
Read the paper in PDF | HTML

An Adaptive Anomaly Detector for Worm Detection
John Mark Agosta, Intel Research; Carlos Diuk-Wasser, Rutgers University; Jaideep Chandrashekar and Carl Livadas, Intel Research
Read the paper in PDF

10:45 a.m.–11:15 a.m.    Break
11:15 a.m.–12:30 p.m.  

Session II

Why Did My PC Suddenly Slow Down?
Sumit Basu, John Dunagan, and Greg Smith, Microsoft Research
Read the paper in PDF | HTML

SPIKE: Best Practice Generation for Storage Area Networks
Prasenjit Sarkar, Ramani Routray, Eric Butler, Chung-hao Tan, and Kaladhar Voruganti, IBM Almaden Research Center; Kiyoung Yang, University of Southern California
Read the paper in PDF | HTML

Analyzing System Logs: A New View of What's Important
Sivan Sabato, Elad Yom-Tov, and Aviad Tsherniak, IBM Haifa Labs; Saharon Rosset, IBM T.J. Watson Research Center
Read the paper in PDF | HTML

Machine Learning Approaches to Network Anomaly Detection
Tarem Ahmed, Boris Oreshkin, and Mark Coates, McGill University
Read the paper in PDF | HTML

12:30 p.m.–2:00 p.m.    Lunch
2:00 p.m.–3:45 p.m.  

Session III

Snitch: Interactive Decision Trees for Troubleshooting Misconfigurations
James Mickens, University of Michigan; Martin Szummer and Dushyanth Narayanan, Microsoft Research, Cambridge
Read the paper in PDF

Fingerpointing Correlated Failures in Replicated Systems
Soila Pertet, Rajeev Gandhi, and Priya Narasimhan, Carnegie Mellon University
Read the paper in PDF | HTML

Structure from Failure
Ralf Herbrich, Thore Graepel, and Brendan Murphy, Microsoft Research, Cambridge
Read the paper in PDF

3:45 p.m.–4:15 p.m.    Break
4:15 p.m.–5:30 p.m.    Panel Discussion
5:30 p.m.–5:45 p.m.    Summary

?Need help? Use our Contacts page.
Last changed: 26 March 2007 jel