|
Published Articles >> Table of Contents >> Abstract
21st International Conference on Data Engineering (ICDE'05)
pp. 978-989
Network-Based Problem Detection for Distributed Systems
Hisashi Kashima, IBM Tokyo Research Laboratory
Tadashi Tsumura, IBM Tokyo Research Laboratory
Tsuyoshi Idé, IBM Tokyo Research Laboratory
Takahide Nogayama, IBM Tokyo Research Laboratory
Ryo Hirade, IBM Tokyo Research Laboratory
Hiroaki Etoh, IBM Tokyo Research Laboratory
Takeshi Fukuda, IBM Tokyo Research Laboratory
Full Article Text:
  
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICDE.2005.93
Send link to a friend
| Abstract |
|
We introduce a network-based problem detection framework
for distributed systems, which includes a data-mining
method for discovering dynamic dependencies among distributed
services from transaction data collected from network,
and a novel problem detection method based on the
discovered dependencies. From observed containments of
transaction execution time periods, we estimate the probabilities
of accidental and non-accidental containments, and
build a competitive model for discovering direct dependencies
by using a model estimation method based on the online
EM algorithm. Utilizing the discovered dependency
information, we also propose a hierarchical problem detection
framework, where microscopic dependency information
is incorporated with a macroscopic anomaly metric
that monitors the behavior of the system as a whole. This
feature is made possible by employing a network-based design
which provides overall information of the system without
any impact on the performance.
|
 Additional Information
|
Citation:
Hisashi Kashima, Tadashi Tsumura, Tsuyoshi Idé, Takahide Nogayama, Ryo Hirade, Hiroaki Etoh, Takeshi Fukuda,
"Network-Based Problem Detection for Distributed Systems,"
icde,
pp. 978-989,
21st International Conference on Data Engineering (ICDE'05),
2005
|
|
