NFV technology is the ideal solution for service providers to deliver reliable and efficient networks and meet urgent needs. NFV deployments face significant challenges such as observability which covers information’s collection related to the platform and applications (metrics). This operation (telemetry) was done by physical probes and separators in the network. However, NFV on a K8s platform has revealed new metrics such as container resource consumption. Telemetry can serve as a source for advanced provisioning and service assurance, enabling not only corrective actions, but also preventive actions against the operation of network services. In addition, data centers require anomaly detection and rapid response to minimize the risk of malfunction. Therefore, it is important to have new collection approaches. This research provides a system capable of collecting new metrics that adopt the concept of realtime delivery and machine learning to derive more data for better visibility into network services. Our telemetry collection system is based on a comparison of technological alternatives for successful telemetry collection, delivery, processing, and visualization. This solution could be tested in a company where the results were promising and positive.
| Date | 28 Apr 2021 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Chamseddine Talhi (Supervisor) |
|---|
Turki, A. (Author),
Talhi (Supervisor),
28 Apr 2021Student thesis: Master's thesis › Master in Engineering: Information Technology Engineering