Execution Anomaly Detection in Distributed Systems
through Unstructured Log Analysis
Qiang FU
1
, Jian-Guang LOU
1
, Yi WANG
2
, Jiang LI
1
1
Microsoft Research Asia
Beijing, P.R. China
{qifu, jlou, jiangli}@microsoft.com
2
Dept. of Computer Science &
Technology
Beijing University of Posts and Telecommunications
Beijing, P.R. China
wangyi.tseg@gmail.com
Abstract -- Detection of execution anomalies is very im-
portant for the maintenance, development, and perfor-
mance refinement of large scale distributed systems. Ex-
ecution anomalies include both work flow errors and low
performance problems. People often use system logs pro-
duced by distributed systems for troubleshooting and
problem diagnosis. However, manually inspecting system
logs to detect anomalies is unfeasible due to the increasing
scale and complexity of distributed systems. Therefore,
