Vitrage is the official OpenStack project dedicated to root cause analysis (RCA) and deduced insights into the behavior of the Cloud. With the help of a graphical model representing the different relationships between system entities, and a friendly UI, Vitrage perfectly organizes, analyzes and visualizes a holistic view of the system.
Fault management in cloud systems is complex. With a mix of virtual & physical resources interacting, problems at one layer can have ripple effects throughout the system, especially when applications can be widely distributed over different physical entities. Achieving clarity of such connections is critical for high-performance systems, such as in NFV
In this session, we will present advanced RCA abilities of Vitrage that tackle real-world problems. With in-depth modeling of the way the virtual and physical entities interact, we will demonstrate how Vitrage traces the impact of problems on the physical layer on virtual machines and applications.
Fault management in cloud systems is complex. With a mix of virtual & physical resources interacting, problems at one layer can have ripple effects throughout the system. This is extremely apparent when trying to connect low-level events, such as NIC failures, with high-level events, such as application problems, especially when applications can be widely distributed over different physical entities. Achieving clarity of such connections is critical for high-performance systems, such as in NFV.
Participants will gain an understanding of how Vitrage, the OpenStack root cause analysis engine, uses its latest data sources to give insights into application-level faults as well as in-depth hardware problems. We will demonstrate how Vitrage provides a clear and holistic view of the system, from physical infrastructure to virtual application, which can then be used to drive application management policy.
