We are seeking someone to join our efforts in the area of automating performance
management of distributed systems. This encompasses detecting, diagnosing,
and acting on availability and performance problems.
We are exploring the application of learning/data mining techniques to several
of these tasks,
including: (a) identifying patterns that are leading indicators of
failures; (b) discovering abstraction hierarchies that characterise problem
causes; and (c) visualising patterns to aid administrators in labelling
problem situations.
The qualified individual should have substantial experience with implementing
machine learning/data mining
algorithms and testing the algorithms on real-world data, especially
system/network performance/management data.
Candidates should have at least BS (MS is preferred) in Computer Science or a
closely related discipline (e.g., EE, Math) .
Knowledge of machine learning, data mining or statistics is desired. Knowledge
of computer networks and distributed
systems is helpful.
If interested, please send your resume to the following address.
Thanks,
Joseph L Hellerstein, Ph.D.
IBM Thomas J. Watson Research Center
30 Sawmill River Road, Rte 9A
Hawthorne, New York
(914) 784-7506