Software-defined cloud interconnect (SDCI) connections are established through different providers with specific or proprietary characteristics. However, no existing scaling technology guarantees the automatic provisioning and teardown of an entire network of circuits, from a software-defined wide area network (SD-WAN) router to the cross connect that is required to link to a cloud provider. Presented herein are techniques that not only abstract different implementations (including hypervisors, software images, releases etc.) but also provide for the ability to detect congestion and implement end-to-end remediation from branch devices to cloud workloads. The presented techniques optimize the costs of resources (such as SD-WAN routers and middle-mile connections) without compromising the level of service that is offered by different middle-mile providers; allow an SDCI’s automatic scaling of VMs and connections to be tied to the specific network SLA requirements of a user-application combination; and support networking solutions that enable a customer to build automated, scalable, and reliable interconnections that deliver richer cloud-based applications and services to enterprise customers while increasing operational efficiency.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Valentini, Giorgio; Subramanyasetty, Pradeep Kanavihalli; Venkatapathy, Venkat; and Cherukuri, Madhavi, "TELEMETRY-BASED RESOURCE SCALING AND TRAFFIC OPTIMIZATION IN CLOUD INTERCONNECTS", Technical Disclosure Commons, (June 25, 2023)