Intent-based network operation is an essential paradigm to create autonomous networks as it eliminates the complexity of manual configurations and replaces it with an abstracted, automated method. To enable intent-based networks that can handle and resolve conflicts under shared and limited network resources, prediction of the impact of proposed configuration and topology changes is crucial. In this respect, Graph Neural Networks (GNN)s are gaining attention in the networking domain thanks to their ability to process data based on close to real modeling of networks as a graph. In this paper, we develop a GNN framework of intent-based networks that takes the network information as inputs and predicts the specific key performance indicators (KPIs) of intents. We propose to make use of GNNs for accurate prediction of the impacts of proposed actions on all of the active KPIs of an intent-based network. Experiments with different configurations are conducted and the empirical evaluations demonstrate that a GNN based prediction agent outperforms a baseline neural-network based agent in both prediction performance and ability to generalize to previously unseen configurations.

The ever-growing complexity in networks and the various future use cases with diverse, and often stringent performance requirements call for a higher level of automation. A tool to achieve this higher level of automation is intent-based management, where the human operator only communicates via high-level intents with the network. Intents are declarative in nature, they specify the desired state but don’t specify how to achieve it. The system will assure that the intents are fulfilled, by monitoring the behavior of the network, and adjusting its configurations if needed. In future networks that will serve multiple tenants and use cases at the same time, many intents will co-exist. As the resources are limited, conflicts may arise between intents. Therefore, a solution is needed for conflict detection and resolution. In this paper, we present a system that is capable of handling multiple intents and detecting and resolving conflicts at run-time. To show the feasibility, we implement our approach in an end-to-end prototype.

Bloom Filter based multicast has been proposed as a source-specific multicast solution to eliminate the multicast state requirements in the routers. However, the inherent limitation, the false positives, in the Bloom filter data structure amplifies the bandwidth wastage when the multicast tree scales to a large number of receivers. In this paper, we propose an algorithm which enhances the performance of the Bloom filter based multicast. It keeps the bandwidth waste below an acceptable upper bound while scaling the multicast tree for a large number of receivers. The large multicast tree is split into multiple smaller ones which are encoded into separate Bloom filters. Our algorithm enables multicast forwarding to be efficient - with careful setting of some parameters - for a hundreds of receivers as compared to the 20-30 receivers per group in the original technique. Furthermore, our algorithm, while slightly increasing the state requirements in the multicast sources, retains the desired property of statelessness in the intermediate routers.

Several recently proposed multicast protocols use in-packet Bloom filters to encode multicast trees. These mechanisms are in principle highly scalable because no per-flow state is required in the routers and because routing decisions can be made efficiently by simply checking for the presence of outbound links in the filter. Yet, the viability of previous approaches is limited by the possibility of forwarding anomalies caused by false positives inherent in Bloom filters. This paper explores such anomalies, namely (1) packets storms, (2) forwarding loops and (3) flow duplication. We propose stateless solutions that increase the robustness and the scalability of Bloom filter-based multicast protocols. In particular, we show that the parameters of the filter need to be varied to guarantee the stability of the packet forwarding, and we present a bit permutation technique that effectively prevents both accidental and maliciously created anomalies. We evaluate our solutions in the context of BloomCast, a source-specific inter-domain multicast protocol, using analytical methods and simulations.

Several publish/subscribe (pub/sub) and data-oriented networking proposals have been presented to overcome limitations of the current message- and host-centric Internet. However, security issues of these solutions have not been addressed comprehensively. In this paper we examine roles of actors comprising an inter-domain pub/sub network, together with security requirements and minimal required trust associations arising from this setting. We then introduce and analyze a security design for a clean-slate pub/sub network architecture that secures both the control and data planes. The solution addresses availability and data integrity while remaining scalable and usable.

A large fraction of today's Internet applications are internally publish/subscribe in nature; the current architecture makes it cumbersome and inept to support them. In essence, supporting efficient publish/subscribe requires data-oriented naming, efficient multicast, and in-network caching. Deployment of native IP-based multicast has failed, and overlay-based multicast systems are inherently inefficient. We surmise that scalable and efficient publish/subscribe will require substantial architectural changes, such as moving from endpoint-oriented systems to information-centric architectures.