arXiv subjects

Jérémie Leguay

Publications and source records attributed to Jérémie Leguay.

At least 19 recordsLinked to original sources

arXiv2025

Atomic Column Generation For Consensus Between Algorithms: Application to Path Computation

In real-life applications, most optimization problems are variants of well-known combinatorial optimization problems, including additional constraints to fit with a particular use case. Usually, efficient algorithms to handle a restricted subset of these additional constraints already exist, or can be easily derived, but combining them together is difficult. The goal of our paper is to provide a framework that allows merging several so-called atomic algorithms to solve an optimization problem including all associated additional constraints together. The core proposal, referred to as Atomic Column Generation (ACG) and derived from Dantzig-Wolfe decomposition, allows converging to an optimal global solution with any kind of atomic algorithms. We show that this decomposition improves the continuous relaxation and describe the associated Branch-and-Price algorithm. We consider a specific use case in telecommunication networks where several Path Computation Elements (PCE) are combined as atomic algorithms to route traffic. We demonstrate the efficiency of ACG on the resource-constrained shortest path problem associated with each PCE and show that it remains competitive with benchmark algorithms.

cs.DM↗

arXiv2024

Safe Load Balancing in Software-Defined-Networking

High performance, reliability and safety are crucial properties of any Software-Defined-Networking (SDN) system. Although the use of Deep Reinforcement Learning (DRL) algorithms has been widely studied to improve performance, their practical applications are still limited as they fail to ensure safe operations in exploration and decision-making. To fill this gap, we explore the design of a Control Barrier Function (CBF) on top of Deep Reinforcement Learning (DRL) algorithms for load-balancing. We show that our DRL-CBF approach is capable of meeting safety requirements during training and testing while achieving near-optimal performance in testing. We provide results using two simulators: a flow-based simulator, which is used for proof-of-concept and benchmarking, and a packet-based simulator that implements real protocols and scheduling. Thanks to the flow-based simulator, we compared the performance against the optimal policy, solving a Non Linear Programming (NLP) problem with the SCIP solver. Furthermore, we showed that pre-trained models in the flow-based simulator, which is faster, can be transferred to the packet simulator, which is slower but more accurate, with some fine-tuning. Overall, the results suggest that near-optimal Quality-of-Service (QoS) performance in terms of end-to-end delay can be achieved while safety requirements related to link capacity constraints are guaranteed. In the packet-based simulator, we also show that our DRL-CBF algorithms outperform non-RL baseline algorithms. When the models are fine-tuned over a few episodes, we achieved smoother QoS and safety in training, and similar performance in testing compared to the case where models have been trained from scratch.

cs.NI↗

arXiv2024

Alternative paths computation for congestion mitigation in segment-routing networks

In backbone networks, it is fundamental to quickly protect traffic against any unexpected event, such as failures or congestions, which may impact Quality of Service (QoS). Standard solutions based on Segment Routing (SR), such as Topology-Independent Loop-Free Alternate (TI-LFA), are used in practice to handle failures, but no distributed solutions exist for distributed and tactical congestion mitigation. A promising approach leveraging SR has been recently proposed to quickly steer traffic away from congested links over alternative paths. As the pre-computation of alternative paths plays a paramount role to efficiently mitigating congestions, we investigate the associated path computation problem aiming at maximizing the amount of traffic that can be rerouted as well as the resilience against any 1-link failure. In particular, we focus on two variants of this problem. First, we maximize the residual flow after all possible failures. We show that the problem is NP-Hard, and we solve it via a Benders decomposition algorithm. Then, to provide a practical and scalable solution, we solve a relaxed variant problem, that maximizes, instead of flow, the number of surviving alternative paths after all possible failures. We provide a polynomial algorithm. Through numerical experiments, we compare the two variants and show that they allow to increase the amount of rerouted traffic and the resiliency of the network after any 1-link failure.

cs.DM↗

arXiv2024

Towards Safe Load Balancing based on Control Barrier Functions and Deep Reinforcement Learning

Deep Reinforcement Learning (DRL) algorithms have recently made significant strides in improving network performance. Nonetheless, their practical use is still limited in the absence of safe exploration and safe decision-making. In the context of commercial solutions, reliable and safe-to-operate systems are of paramount importance. Taking this problem into account, we propose a safe learning-based load balancing algorithm for Software Defined-Wide Area Network (SD-WAN), which is empowered by Deep Reinforcement Learning (DRL) combined with a Control Barrier Function (CBF). It safely projects unsafe actions into feasible ones during both training and testing, and it guides learning towards safe policies. We successfully implemented the solution on GPU to accelerate training by approximately 110x times and achieve model updates for on-policy methods within a few seconds, making the solution practical. We show that our approach delivers near-optimal Quality-of-Service (QoS performance in terms of end-to-end delay while respecting safety requirements related to link capacity constraints. We also demonstrated that on-policy learning based on Proximal Policy Optimization (PPO) performs better than off-policy learning with Deep Deterministic Policy Gradient (DDPG) when both are combined with a CBF for safe load balancing.

cs.NI↗

arXiv2024

Virtual Multi-Topology Routing for QoS Constraints

Multi-topology routing (MTR) provides an attractive alternative to segment routing for traffic engineering when network devices cannot be upgraded. However, due to a high overhead in terms of link state messages exchanged by topologies and the need to frequently update link weights to follow evolving network conditions, MTR is often limited to a small number of topologies and the satisfaction of loose QoS constraints. To overcome these limitations we propose vMTR, an MTR extension where demands are routed over virtual topologies that are silent, i.e., they do not exchange LSA messages, and that are continuously derived from a very limited set of real topologies, optimizing each a QoS parameter. In this context, we present a polynomial and exact algorithm for vMTR and, as a benchmark, a local search algorithm for MTR. We show that vMTR helps reducing drastically the number of real topologies and that it is more robust to QoS changes.

cs.NI↗

arXiv2023

Global QoS Policy Optimization in SD-WAN

In modern SD-WAN networks, a global controller is able to steer traffic on different paths based on application requirements and global intents. However, existing solutions cannot dynamically tune the way bandwidth is shared between flows inside each overlay link, in particular when the available capacity is uncertain due to cross traffic. In this context, we propose a global QoS (Quality of Service) policy optimization model that dynamically adjusts rate limits of applications based on their requirements to follow the evolution of network conditions. It relies on a novel cross-traffic estimator for the available bandwidth of overlay links that only exploits already available measurements. We propose two local search algorithms, one centralized and one distributed, that leverage cross-traffic estimation. We show in packet-level simulations a significant performance improvement in terms of SLA (Service Level Agreement) satisfaction. For instance, the adaptive tuning of load balancing and QoS policies based on cross-traffic estimation can improve SLA satisfaction by $40\%$ compared to static policies.

cs.NI↗

arXiv2023

Protected load balancing problem: Neural-network based approximation for non-convex optimization

Nowadays, centralized Path Computation Elements (PCE) integrate control plane algorithms to optimize routing and load-balancing continuously. When a link fails, the traffic load is automatically transferred to the remaining paths according to the configuration of load-balancers. In this context, we propose a load-balancing method that anticipates load transfers to ensure the protection of traffic against any Shared Risk Link Group (SRLG) failure. The main objective of this approach is to make better use of bandwidth compared to existing methods. It consists in reserving a minimum amount of extra bandwidth on links so that the rerouting of traffic is guaranteed. We propose a non-linear non-convex model for the problem of minimizing the bandwidth reservation cost. We introduce a new approximation approach based on a neural network to convexify the problem and apply Kelley's cutting plane method to solve the problem. Finally, we show that our algorithm significantly improves the CPU time against a compact model solved using the SCIP solver.

cs.NI↗

arXiv2022

Routing and QoS Policy Optimization in SD-WAN

In modern SD-WAN networks, a global controller continuously optimizes application and user intents by selecting the proper routing policies for each application. Nevertheless, the competition between flows can still occur at each overlay link and it may degrade Quality of Service (QoS). To mitigate performance degradations in case of congestion, QoS policies can also be dynamically optimized to limit the rate of low-priority flows and share the bandwidth among all flows in a fair manner. This demonstration presents a comprehensive control plane architecture to coordinate path selection and rate allocation, in order to meet application requirements while optimizing a global objective (e.g., low congestion, best quality, and minimum cost).

cs.NI↗

arXiv2022

Scalable Damper-based Deterministic Networking

With 5G networking, deterministic guarantees are emerging as a key enabler. In this context, we present a scalable Damper-based architecture for Large-scale Deterministic IP Networks (D-LDN) that meets required bounds on end-to-end delay and jitter. This work extends the original LDN architecture, where flows are shaped at ingress gateways and scheduled for transmission at each link using an asynchronous and cyclic opening of gate-controlled queues. To further relax the need for clock synchronization between devices, we use dampers, that consist in jitter regulators, to control the burstiness flows to provide a constant target delay at each hop. We introduce in details how data plane functionalities are implemented at all nodes (gateways and core) and we derive how the end-to-end delay and jitter are calculated. For the control plane, we propose a column generation algorithm to quickly take admission control decisions and maximize the accepted throughput. For a set of flows, it determines acceptance and selects the best shaping and routing policy. Through a proof-of-concept implementation in simulation, we verify that the architecture meets promised guarantees and that the control plane can operate efficiently at large-scale.

cs.NI↗

arXiv2022

Graph Convolutional Reinforcement Learning for Collaborative Queuing Agents

In this paper, we explore the use of multi-agent deep learning as well as learning to cooperate principles to meet stringent service level agreements, in terms of throughput and end-to-end delay, for a set of classified network flows. We consider agents built on top of a weighted fair queuing algorithm that continuously set weights for three flow groups: gold, silver, and bronze. We rely on a novel graph-convolution based, multi-agent reinforcement learning approach known as DGN. As benchmarks, we propose centralized and distributed deep Q-network approaches and evaluate their performances in different network, traffic, and routing scenarios, highlighting the effectiveness of our proposals and the importance of agent cooperation. We show that our DGN-based approach meets stringent throughput and delay requirements across all scenarios.

cs.NI↗

arXiv2018

Multi-Path Alpha-Fair Resource Allocation at Scale in Distributed Software Defined Networks

The performance of computer networks relies on how bandwidth is shared among different flows. Fair resource allocation is a challenging problem particularly when the flows evolve over time. To address this issue, bandwidth sharing techniques that quickly react to the traffic fluctuations are of interest, especially in large scale settings with hundreds of nodes and thousands of flows. In this context, we propose a distributed algorithm based on the Alternating Direction Method of Multipliers (ADMM) that tackles the multi-path fair resource allocation problem in a distributed SDN control architecture. Our ADMM-based algorithm continuously generates a sequence of resource allocation solutions converging to the fair allocation while always remaining feasible, a property that standard primal-dual decomposition methods often lack. Thanks to the distribution of all computer intensive operations, we demonstrate that we can handle large instances at scale.

cs.NI↗

arXiv2018

Lower Bounds for the Fair Resource Allocation Problem

The $α$-fair resource allocation problem has received remarkable attention and has been studied in numerous application fields. Several algorithms have been proposed in the context of $α$-fair resource sharing to distributively compute its value. However, little work has been done on its structural properties. In this work, we present a lower bound for the optimal solution of the weighted $α$-fair resource allocation problem and compare it with existing propositions in the literature. Our derivations rely on a localization property verified by optimization problems with separable objective that permit one to better exploit their local structures. We give a local version of the well-known midpoint domination axiom used to axiomatically build the Nash Bargaining Solution (or proportionally fair resource allocation problem). Moreover, we show how our lower bound can improve the performances of a distributed algorithm based on the Alternating Directions Method of Multipliers (ADMM). The evaluation of the algorithm shows that our lower bound can considerably reduce its convergence time up to two orders of magnitude compared to when the bound is not used at all or is simply looser.

cs.NI↗

arXiv2017

Real-Time Fair Resource Allocation in Distributed Software Defined Networks

The performance of computer networks relies on how bandwidth is shared among different flows. Fair resource allocation is a challenging problem particularly when the flows evolve over time.To address this issue, bandwidth sharing techniques that quickly react to the traffic fluctuations are of interest, especially in large scale settings with hundreds of nodes and thousands of flows. In this context, we propose a distributed algorithm that tackles the fair resource allocation problem in a distributed SDN control architecture. Our algorithm continuously generates a sequence of resource allocation solutions converging to the fair allocation while always remaining feasible, a property that standard primal-dual decomposition methods often lack. Thanks to the distribution of all computer intensive operations, we demonstrate that we can handle large instances in real-time.

cs.NI↗

arXiv2017

Virtual Function Placement for Service Chaining with Partial Orders and Anti-Affinity Rules

Software Defined Networking and Network Function Virtualization are two paradigms that offer flexible software-based network management. Service providers are instantiating Virtualized Network Functions - e.g., firewalls, DPIs, gateways - to highly facilitate the deployment and reconfiguration of network services with reduced time-to-value. They employ Service Function Chaining technologies to dynamically reconfigure network paths traversing physical and virtual network functions. Providing a cost-efficient virtual function deployment over the network for a set of service chains is a key technical challenge for service providers, and this problem has recently caught much attention from both Industry and Academia. In this paper, we propose a formulation of this problem as an Integer Linear Program that allows one to find the best feasible paths and virtual function placement for a set of services with respect to a total financial cost, while taking into account the (total or partial) order constraints for Service Function Chains of each service and other constraints such as end-to-end latency, anti-affinity rules between network functions on the same physical node and resource limitations in terms of network and processing capacities. Furthermore, we propose a heuristic algorithm based on a linear relaxation of the problem that performs close to optimum for large scale instances.

cs.NI↗

arXiv2017

Overlay Routing for Fast Video Transfers in CDN

Content Delivery Networks (CDN) are witnessing the outburst of video streaming (e.g., personal live streaming or Video-on-Demand) where the video content, produced or accessed by mobile phones, must be quickly transferred from a point to another of the network. Whenever a user requests a video not directly available at the edge server, the CDN network must 1) identify the best location in the network where the content is stored, 2) set up a connection and 3) deliver the video as quickly as possible. For this reason, existing CDNs are adopting an overlay structure to reduce latency, leveraging the flexibility introduced by the Software Defined Networking (SDN) paradigm. In order to guarantee a satisfactory Quality of Experience (QoE) to users, the connection must respect several Quality of Service (QoS) constraints. In this paper, we focus on the sub-problem 2), by presenting an approach to efficiently compute and maintain paths in the overlay network. Our approach allows to speed up the transfer of video segments by finding minimum delay overlay paths under constraints on hop count, jitter, packet loss and relay processing capacity. The proposed algorithm provides a near-optimal solution, while drastically reducing the execution time. We show on traces collected in a real CDN that our solution allows to maximize the number of fast video transfers.

cs.NI↗

arXiv2016

Online and Global Network Optimization: Towards the Next-Generation of Routing Platforms

The computation power of SDN controllers fosters the development of a new generation of control plane that uses compute-intensive operations to automate and optimize the network configuration across layers. From now on, cutting-edge optimization and machine learning algorithms can be used to control networks in real-time. This formidable opportunity transforms the way routing systems should be conceived and designed. This paper presents a candidate architecture for the next generation of routing platforms built on three main pillars for admission control, re-routing and monitoring that would have not been possible in legacy control planes.

cs.NI↗

arXiv2015

Adapting Caching to Audience Retention Rate: Which Video Chunk to Store?

Rarely do users watch online contents entirely. We study how to take this into account to improve the performance of cache systems for video-on-demand and video-sharing platforms in terms of traffic reduction on the core network. We exploit the notion of "Audience retention rate", introduced by mainstream online content platforms and measuring the popularity of different parts of the same video content. We first characterize the performance limits of a cache able to store parts of videos, when the popularity and the audience retention rate of each video are available to the cache manager. We then relax the assumption of known popularity and we propose a LRU (Least Recently Used) cache replacement policy that operates on the first chunks of each video. We characterize its performance by extending the well-known Che's approximation to this case. We prove that, by refining the chunk granularity, the chunk-LRU policy increases its performance. It is shown numerically that even for a small number of chunks (N=20), the gains of chunk-LRU are still significant in comparison to standard LRU policy that caches entire files, and they are almost optimal.

cs.NI↗

arXiv2015

TACTICS: TACTICal Service Oriented Architecture

Due to the increasing complexity and heterogeneity of contemporary Command, Control, Communications, Computers, & Intelligence systems at all levels within military organizations, the adoption of the Service Oriented Architectures (SOA) principles and concepts is becoming essential. SOA provides flexibility and interoperability of services enabling the realization of efficient and modular information infrastructure for command and control systems. However, within a tactical domain, the presence of potentially highly mobile actors equipped with constrained communications media (i.e., unreliable radio networks with limited bandwidth) limits the applicability of traditional SOA technologies. The TACTICS project aims at the definition and experimental demonstration of a Tactical Services Infrastructure enabling tactical radio networks (without any modifications of the radio part of those networks) to participate in SOA infrastructures and provide, as well as consume, services to and from the strategic domain independently of the user's location.

cs.SE↗