logs2graphs: Data-driven graph representation and visualization of log data
Date Issued
2022
Author(s)
Andonov, Stefan
Jovev, Viktor
Kitanovski, Aleksandar
Krsteski, Aleksandar
Madjarov, Gjorgji
Abstract
In recent years, AIOps has helped a lot with the
exploration of different types of resources, in the processes of
optimization and automation of complex IT operations. One of
the main resources that AIOps is exploring is system logs. There
are many techniques based on machine learning in AIOps that
help in logs anomaly detection, logs prediction, and root cause
analysis guided by logs, but a majority of them are considering
log messages either individually or as log sequences, without
exploring the relationships between different types of logs. We
believe that those relationships can be expressed via using graph
representations of log messages and those representations can
be utilized in almost any AIOps operation. Therefore in this
paper, we present logs2graphs, an open-source system for the
creation and visualization of such graph representations of log
messages, which is compatible with several publicly available
log sources and expandable to other log sources.
exploration of different types of resources, in the processes of
optimization and automation of complex IT operations. One of
the main resources that AIOps is exploring is system logs. There
are many techniques based on machine learning in AIOps that
help in logs anomaly detection, logs prediction, and root cause
analysis guided by logs, but a majority of them are considering
log messages either individually or as log sequences, without
exploring the relationships between different types of logs. We
believe that those relationships can be expressed via using graph
representations of log messages and those representations can
be utilized in almost any AIOps operation. Therefore in this
paper, we present logs2graphs, an open-source system for the
creation and visualization of such graph representations of log
messages, which is compatible with several publicly available
log sources and expandable to other log sources.
Subjects
File(s)![Thumbnail Image]()
Loading...
Name
CIIT_2022_16.pdf
Size
1.48 MB
Format
Adobe PDF
Checksum
(MD5):c0ac370cc1a27ddf16a02bea93215039