L’ INGENIERIE DU TRAFIC DANS LES RESEAUX MPLS
L’ingénierie du trafic
Traffic engineering (TE) is a process that specifies how traffic is managed within a given network, primarily focusing on optimizing the performance of operational networks It involves the application of technologies and scientific principles for measuring, modeling, characterizing, and controlling Internet traffic to achieve specific performance objectives In MPLS networks, the key aspects of interest in TE are measurement and control.
The primary performance objectives associated with Traffic Engineering (TE) can be classified into two categories: traffic-oriented and resource-oriented goals Traffic-oriented objectives focus on enhancing the Quality of Service (QoS) for data flows, which includes minimizing packet loss and delay, as well as maximizing throughput On the other hand, resource-oriented objectives aim to optimize resource utilization, with bandwidth management being a critical factor Key goals in this category include minimizing congestion and achieving effective load balancing.
MPLS et la TE
MPLS is highly significant for traffic engineering (TE) as it integrates many valuable features of overlay models, such as IP over ATM and IP over Frame Relay, at a lower cost compared to other solutions It also enables the automation of various TE functions, enhancing efficiency and performance.
Explicit LSPs that are not bound by destination-based transfer paradigms can be created manually by network operators or automatically through protocols These LSPs can be maintained efficiently, allowing traffic trunks to be instantiated and mapped onto them Additionally, a set of attributes can be associated with traffic trunks to modulate their behavioral characteristics, while another set of attributes can be linked to resources that constrain the placement of LSPs and the traffic trunks traversing these LSPs.
However, there are challenges in implementing Traffic Engineering (TE) on MPLS The three fundamental issues of TE on MPLS include: first, how to map packets onto Forwarding Equivalence Classes (FEC); and second, how to map FEC onto traffic trunks.
Thirdly, how do we map traffic trunks onto the physical topology of a network through LSPs? Several studies have addressed this issue, focusing on the integration of DiffServ techniques with MPLS infrastructure, as well as the development of routing protocols and label distribution mechanisms.
Integrating the DiffServ technique into an MPLS network offers significant advantages by allowing the specification of routing paths and packet behaviors (PHB) within router queues This is achieved by classifying FECs according to DiffServ service classes, such as Expedited Forwarding (EF).
AFxy, BE) Il y a deux faỗons de projeter le DSCP de DiffServ sur les ộtiquettes de MPLS : E-LSP
(Exp-inferred LSP) et L-LSP (Label only inferred LSP) [28, 29] Les étiquettes seront assignés et distribuées aux routeurs grâce à un LDP
One of the intriguing aspects of TE in MPLS is the dynamic and automatic management of access to an MPLS-DiffServ network This involves allocating external traffic flows to various LSPs based on the negotiated Service Level Specification (SLS), application characteristics, and potentially user profiles For instance, when a voice application is introduced, determining the appropriate LSP for its allocation is crucial to ensure a high level of service quality.
Typically, traffic is routed either statically by the operator or through a label routing and distribution protocol Before assigning a flow to a pipeline, it is essential to implement a Call Admission Control (CAC) procedure to ensure Quality of Service (QoS) compliance To incorporate artificial intelligence into telecommunications, we propose an approach utilizing intelligent agents for managing access within an MPLS-DiffServ network.
L’ OBJECTIF DU PROJET
The goal of the project is to design and develop simulation techniques and software for MPLS-DiffServ networks that incorporate intelligent agents In this context, I collaborated with another intern who focused on the modeling and development of an MPLS-DiffServ network.
This internship focuses on developing a generic agent model for selecting LSPs and evaluating its effectiveness along with its drawbacks in terms of TE We aim to apply agent-based artificial intelligence to enhance traffic engineering in an MPLS network.
DiffServ Nous creusons l’aspect du choix des LSP en satisfaisant la QoS demandée et l’équilibrage de la charge du réseau Pour cela, on tente de répondre aux questions suivantes : (i)
Pourquoi et comment introduit-on l’agent intelligent à un réseau MPLS? (ii) Quels sont ses avantages et inconvénients ? (iii) Comment affecte-t-on les flux de paquets à les divers LSP? (iv)
What parameters can the agent manipulate? What is the architecture of the agents? How do we assess the effectiveness of the model? These questions will be explored in the remainder of the document.
Cette mémoire est organisée comme suit La section 2 présente certaines applications des agents intelligents dans le domaine de la télécommunication, la réalisation de réseaux MPLS-
DiffServ et l’environnement de simulation J-Sim Notre approche de modélisation d’un réseau
Section 3 will present MPLS-DiffServ and the intelligent agent-based network access management model In Section 4, we will detail the simulation parameters and the results obtained Finally, Section 5 will conclude this paper and outline our future work.
Dans cette partie, nous allons aborder l’utilisation des agents dans le domaine de la télécommunication et un environnement particulier de simulation d’un réseau de télécommunications
2.1 L’agent intelligent dans les réseaux de Télécommunications
The complexity and dynamism of telecommunications networks make their management and control increasingly challenging The implementation of intelligent agents in this field minimizes human intervention and enhances network performance These agents can identify situations and offer improved solutions, streamlining operations and optimizing efficiency.
Gạti describes a model of agents designed for congestion detection, traffic management, and dynamic threshold adjustment of congestion control mechanisms in ATM networks Each agent operates with its own knowledge and is assigned to each network node, allowing for independent operation when necessary The agent architecture is based on blackboard principles, consisting of three components: a knowledge module, a control module, and a communication module This work has significantly contributed to the successful application of Distributed Artificial Intelligence (IAD) in the telecommunications field.
Legge [2] proposes the development of an Autonomous Agent-Based Management System for AMT networks, where each network node operates its own society of agents Each agent is responsible for managing a layer of the ISO reference model, providing defined interfaces and abstraction levels to facilitate the development of interoperable communication protocols However, the implementation of three lower layers is still pending: the Switch Control Agent for the physical layer, the Neighbour Discovery Agent for the link layer, and the Network Agent for the network layer Additionally, there is a specialized agent called the Inter-Society Communication Agent (ISCA) that maintains a connection to the ATM switch and communicates with other agents.
Vilà presents a multi-agent system designed for the dynamic configuration of network resources, utilizing bandwidth reservation and route restoration mechanisms This system effectively manages virtual networks, such as virtual routes in ATM networks or Label Switched Paths (LSP) in MPLS networks It features two types of agents: M-agents, which are reactive and responsible for monitoring the state of virtual routes and detecting congestion, and P-agents, which are deliberative agents present at each node to optimize network performance.
A promising approach to network control and routing involves the use of mobile agents Vittori introduces an intelligent routing algorithm known as Q-agents, which operates based on a set of actions influenced by the agent's interaction environment This algorithm integrates three learning strategies: Q-learning, dual reinforcement learning, and ant colony behavior A group of agents navigates the network independently and concurrently to identify optimal routes These agents share their knowledge of route quality through indirect communication, enhancing overall routing efficiency.
The article discusses a novel approach to modeling and simulating active networks through a behavioral multi-agent system designed to manage network dynamics Agents, characterized by properties such as autonomy, sociability, communication, cooperation, and learning, are employed to operate within active network environments They perform dynamic and intelligent control to prevent congestion and packet loss Each node, regarded as an active entity, is represented by a multi-agent system and exhibits various behaviors—basic, selective, cautious, loyal, and unfaithful—when processing packets across three different quality of service classes This behavioral modeling aims to reflect the actions of each entity, resulting in significant improvements in service quality, including reduced packet loss and response times.
Ces travaux ont prouvé l’efficacité des agents intelligents pour la gestion dynamique de réseaux de télécommunications
Les concepts et l’architecture de routage d’un réseau MPLS-DiffServ ont été présentés dans
There are two fundamental issues in implementing MPLS-DiffServ networks: the DSCP is included in the IP packet header, while LSRs only examine frame labels; additionally, the DSCP field has 6 bits, whereas the experimental field (EXP) in labels has only 3 bits To address these challenges, two solutions, E-LSP and L-LSP, have been proposed The E-LSP solution employs a method of projecting multiple Behavior Aggregates (BAs) onto a single LSP, allowing all packets belonging to these BAs to share the same label.
Le champ EXP de l’en-tête MPLS est utilisé pour spécifier le PHB applicable à chaque paquet
Ce PHB comprends les paramètres de la préférence de rejet et d’ordonnancement La deuxième projette un BA singulier sur un LSP, les DSCP sont encodés implicitement dans les étiquettes
Par conséquent, le champ EXP devrait être utilisé pour le codage de la préférence de rejet des paquets
En ce qui concerne la réalisation de la TE dans un réseau MPLS-DiffServ, le projet
TEQUILA (Traffic Engineering for Quality of Service in the Internet at Large Scale) aims to study, specify, implement, and validate a comprehensive set of definitions for services and tools in traffic engineering Its goal is to ensure quantitative end-to-end Quality of Service (QoS) through network dimensioning, admission control, and dynamic resource management within DiffServ networks based on MPLS infrastructure Currently, the project focuses on developing a prototype of the DiffServ model on the MPLS framework.
Dans le cadre de ce stage, on utilise l’environnement de simulation J-Sim (Java Simulation) pour créer et expérimenter des simulations d’un réseau MPLS-DiffServ
J-Sim is a software package designed for simulating discrete-time systems in Java It offers a development environment based on the Autonomous Component Architecture (ACA), where all elements are treated as components These components communicate and exchange data through their ports, facilitating efficient system modeling and simulation.
Fig 1 L’architecture des composants de base
J-Sim has developed an abstract network model known as the INET Network Model, where all entities within a network are considered components These components include networks, nodes, network interface cards, and physical links, which together form a telecommunications network A network itself is a composite component made up of nodes, links, and smaller networks Each node is a distinct component that encompasses applications, protocol modules, and a core service layer (CSL) The figure below illustrates the INET network modeling framework.
Fig 2 Le modèle de réseau INET
The following figure illustrates the internal structure of a node within the abstract model Unlike the TCP/IP model, which consists of four layers, or the OSI reference model with seven layers, this model simplifies the architecture by incorporating only two layers At the top, there is a protocol layer.
UPL(Upper protocol layer) et celle de service du noyau CSL
Fig 3 La structure interne d’un nœud de réseau
La décomposition du CSL sera montrée en détails dans l’annexe concernant J-Sim
2.3.3 Modélisation du réseau MPLS dans J-Sim