uABNO: A Cloud-Native Architecture for Optical SDN Controllers

Abstract

We present a cloud-native architecture for Optical SDN Controllers based on ABNO architecture and gRPC interfaces, which is demonstrated and evaluated. Autoscaling mechanisms for high request loads and auto-healing support are evaluated. © 2020 The Author(s) 1. Introduction Software Defined Networking (SDN) is a consolidated network architecture paradigm that provides network programmability by separating the control plane logic from the data plane forwarding infrastructure. This decoupling, provides novel benefits to network operators, such as CAPEX savings by replacing dedicated hardware network equipment by software-driven network elements and OPEX thanks to faster new service introduction. SDN has been accompanied with new open standard interfaces, such as NETCONF, gRPC, OpenFlow or P4, which allow to interact with the network elements from centralized entities generally defined as SDN controllers [1]. SDN controllers currently are developed as single monolithic and resource-hungry applications, which might be replicated in case of need for resiliency or extra resources. This leads to non-efficient use of the resources, does not provide scaling mechanisms for high loads of connectivity service requests, and incur on extra delays in each request, not allowing a cloud-scale number of requests. Some network vendors are experimenting with solutions that export the complexity towards applications, but are not dealing with the need to break the monolithic SDN controllers [2]. The Applications-Based Network Operations (ABNO) framework [3] has been standardized by the IETF and it is based on standard protocols and components to efficiently provide a solution to the transport network orchestration. Microservices are a software development technique that structures an application as a collection of interconnected and related services. In a microservices architecture, services are simple and detailed and the protocols are lightweight. For example, gRPC Remote Procedure Calls (RPC) [4] is a protocol designed for cloud native high-performance RPC. It uses HTTP/2 as a transport protocol and uses protocol buffers encodings for transported messages. gRPC has been proven as useful in telemetry, due to its low latency and small byte overhead. In this paper, we propose the application of microservices architecture to the development of an SDN controller. The internal architecture of SDN controller makes it clearly a good candidate for splitting it into microservices, which provide resiliency and scalability features per design. Using ABNO components, several microservices will be defined, including path computation service, NBI service, connectivity service, connection service, topology service, context service, VNTM service, transceiver service, monitoring service, and plugin services. Each microservice will interact with each other using protocol buffers and gRPC interfaces. A common gRPC interface will be provided for health check of the microservices. This novel cloud-native architecture named uABNO (Fig. 1, left) provides ...

First Scalable Machine Learning Based Architecture for Cloud-native Transport SDN Controller

“…Software-Defined Networking (SDN) has enabled the centralized use of multi-layer, multi-domain and multivendor network equipment using open standard Application Programming Interfaces (APIs) instead of distributed managing through proprietary interfaces. This new centralized paradigm needs a new kind of SDN controller architecture able to deal with the aggregation of several transport layers and tens/hundreds of domains, while at the same time being resilient and capable of scaling to a cloud-level number of connections for provisioning and telemetry [1]. Cloud-native architectures are based on micro-services, a set of loosely coupled services that are connected by means of lightweight protocols, such as gRPC Remote Procedure Calls (RPC).…”

“…To address this issues, we extend a cloud-native SDN controller called uABNO [1] with a scalable machine learning framework that would enable multiple autonomic functions, such as the QoT predictor presented in this paper. The auto-scaling properties will also be demonstrated for the machine learning framework in a real WDM/SDM testbed using Docker containers orchestrated and scaled using Kubernetes.…”

Scalable Physical Layer Security Components for Microservice-Based Optical SDN Controllers

“…Fig. 1 shows the architecture of uABNO [3] that implements a cloud-native optical SDN controller. uABNO exposes network topology and CS to external applications (e.g., Operations Support Systems/Business Support Systems -OSS/BSS).…”

“…In this work, we propose and demonstrate a set of microservice-based components that perform physical layer security assessment and mitigation operations in an optical transport network infrastructure. With this purpose in mind, we developed three components able to perform attack inference, detection, and mitigation, and integrated them into uABNO [3] , a microservice-based optical SDN controller. We tested the scalability performance of these Physical Layer Security Components (PLSCs) by measuring their response time 978-1-6654-3868-1/21/$31.00 ©2021 IEEE under different network load conditions.…”

Cloud-native SDN Controller Based on Micro-Services for Transport Networks

“…In [3], the authors have presented the basis for such new generation of SDN controller, which is more flexible and scalable. The proposed novel cloud-native SDN controller is based on IETF Application-Based Network Operations [4], which we refer as µABNO (micro-ABNO).…”