Many new applications/services/microservices are hosted in some form of a cloud network, such as a private, public, or hybrid cloud. This means the network connecting applications/services/microservices to clients is critical infrastructure that is needed to facilitate a good user-experience. However, some cloud network environments that utilize datacenter fabrics may have limited visibility with regard to observability and/or awareness of different network events (e.g., failures, service faults, potential performance optimizations, etc.). Provided herein are techniques through which Service-Level Agreement (SLA) decisions that may be made by third-party tools can be consumed in order to drive flow steering in a datacenter fabric utilizing local, hardware-assisted switch mechanisms. Such techniques may provide for the ability to replace/assist tool/controller interactions by offloading control plane changes for scenarios involving SLA violations, which may result in faster SLA resolution utilizing hardware-assisted switching.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Iyengar, Praveen Parthasarathy; Saxena, Maneesh; Katti, Pralhad; and Rajurwar, Alhad, "TECHNIQUES TO FACILITATE OBSERVABILITY-AWARE AND SELF-HEALING TRAFFIC ENGINEERING", Technical Disclosure Commons, (March 13, 2023)