Nokia develops Vitrage root cause analysis for OpenStack
<iframe src="https://www.youtube.com/embed/R65F57nxF8w?modestbranding=1&rel=0" width="970" height="546" frameborder="0" scrolling="auto" allowfullscreen></iframe>
Elisha Rosensweig, PHD, Director R&D, CBIS, CloudBand, Applications and Analytics, Nokia & Ifat Afek, System Architect, CBIS, CloudBand, Applications and Analytics, Nokia.
Root cause analysis of network faults has traditionally been a manual process. A fault is detected by the monitoring system, the network management system takes remedial action (re-routes traffic or boots up the faulty process on a redundant box - whatever the recovery procedure is) and the network sails on. But it’s preferable to know what’s happened in the network so it can be prevented from happening again, perhaps with worse consequence. This is root cause analysis and it used to mean getting the experts together to try and work out what had happened - after the event.
Automating root cause analysis has been an established application within Nokia’s Cloudband, but nothing like it was being attempted for OpenStack, so Nokia decided to do it with a new product called Vitrage. Nokia's Elisha Rosensweig and Ifat Afek tell Martyn Warwick how the project is proceeding and why root cause analysis is going to be a very handy cloud tool as cloud complexity grows. What comes next? Hint: the goal is to have enough data to analyse to perform root cause 'prediction'.
FILMED AT: OpenStack Summit 2016, Barcelona, Spain