Open Networking Summit Europe 2019

Various studies indicate platform components like memory are prone to errors over prolonged usage leading to huge unpredictable downtime. A resilient & self-managed 5G cloud requires ability of fault detection, fault isolation and re-provisioning of network slices based on hardware reliability errors encountered using open industry standard interfaces to guarantee five nines VNF operation. It is not just sufficient to recover fast from failure, but to close the loop by repairing what is broken and ensuring the service availability.

Our ongoing work in newly formed OPNFV Closed Loop Automation working group will showcase how you can have it all with a demo, with Collectd, Prometheus & OpenStack components Vitrage & Fenix along with available resiliency features that help predict component reliability.