Open Networking Summit Europe 2019

With containers becoming the desired deployment model for a wide range of workloads such as Big Data, IoT, storage, AI, ML, SDN and NFV solutions, Container orchestration tools like Kubernetes, need to evolve to meet stringent networking and resource management requirements. Service Level Objectives for many such workloads demand network I/O performance and capabilities beyond pure connectivity and reachability. Automated actions based on policies as well as predictive and reactive closed loops are becoming an indispensable requirement in the orchestration layer in order to achieve operational Efficiency, better Service performance, reduction in CapEx, OpEx and improved customer experience. This talk focuses on kubernetes scheduling capabilities enabling intelligent placement of workloads as well as for performing automated actions by using system telemetry obtained from Prometheus.