Halina Tarasiuk

EuQoS: End-to-End Quality of Service over Heterogeneous Networks

The EuQoS (End-to-End QoS over Heterogeneous Networks) IST Integrated European Project aimed to define a Next Generation Network architecture that builds, uses and manages end-to-end QoS across different administrative domains and heterogeneous networks (UMTS, xDSL, Ethernet, WiFi, Satellite and IP/ MPLS). The EuQoS architecture preserves the openness and the decentralized decision model of the actual Internet, runs on off-the-shelf hardware and network equipment, and allows end users to request various services without changing the Application Signaling protocol, while meeting regulators’ and users’ Net Neutrality requirements. This paper presents the key elements of the EuQoS architecture and describes the main results obtained in field trials performed on a fully-functional EuQoS system prototype developed over a pan-European testbed. Furthermore, the paper discusses the main strengths of the system and the issues related to its actually deployment on a large scale, from both technical and market points

Provision of end-to-end QoS in heterogeneous multi-domain networks

In this paper, we present the framework to provision end-to-end QoS in heterogeneous multi-domain networks that was implemented in EuQoS system and tested in Pan-European research network. It assumes that a pair of end users, possibly attached to different access networks as xDSL, UMTS, LAN/Ethernet, WiFi, MPLS or Satellite, may choose for its connection an appropriate end-to-end Class of Service, depending on the application they use, e.g. VoIP, VoD, FTP, etc. Each end-to-end Class of Service has its own traffic control mechanisms and algorithms and, as a result, it has the ability to handle traffic streams with assumed guarantees for packet transfer characteristics expressed in the form of loss ratio, mean delay and delay variation. The end-to-end Classes of Service are supported in all the domains (including inter-domain links) independently using specialised inter-domain Class of Service-aware QoS routing protocol which establishes the end-to-end QoS paths. This paper describes the solution and includes exemplary simulation and experimental results.

Traffic Handling in AQUILA QoS IP Network

The paper describes the traffic handling mechanisms implemented in the AQUILA pilot QoS IP network [10]. The AQUILA project enhances the DiffServ architecture concept [1,2,3] by adding new functionality for admission control and resource management as well as by defining new set of network services. Each network service is optimised for specific type of traffic (e.g. nonreactive and reactive) and has its own traffic handling mechanisms. Exemplary measurement results verifying the effectiveness of AQUILA approach for providing QoS are also included.

AC algorithms in AQUILA QoS IP network

This paper presents the admission control (AC) algorithms that are implemented in the prototype multi-service AQUILA QoS IP network [4]. The discussed algorithms were developed for regulating traffic submitted to network services (NS) dedicated for handling: (1) real-time streaming traffic of variable bit rate (VBR) type and (2) elastic traffic, produced by greedy TCP sources. For the former NS, named Premium VBR, the measurement-based AC (MBAC) scheme is proposed. The applied approach employs the well-known Hoeffding bound formula [2] for calculating required link capacity to assure packet loss ratio at a predefined level. Moreover, this method is supported by the declarations about the peak bit rate as well as by the measurements of mean bit rate on aggregate flow level. For the latter NS, named premium multimedia, two alternative AC algorithms are investigated, both designed for assuring requested TCP throughput. While the first of them follows the token bucket marking (TBM) concept [12], the second one adjusts the advertised TCP window size to enable an ideal TCP behaviour (i.e. lossless packet transfer). The simulation results are included for illustrating the advantages of the discussed algorithms.

A framework for providing differentiated QoS guarantees in IP-based network

The paper describes the traffic handling mechanisms implemented in the AQUILA pilot QoS IP network [AQUILA Project Consortium (2001)] [11]. The AQUILA architecture enhances the DiffServ concept [A Conceptual Model for DiffServ Routers (2000), An Architecture for Differentiated Services (1998), An Expedited Forwarding PHB (2001)] by adding new functionality for admission control and resource management as well as by defining new set of Network Services (NSs) [4,6,3]. Each NS is optimised for specific type of traffic (e.g. reactive and non-reactive) and has its own traffic handling mechanisms. The mentioned mechanisms operate at different time scales, ranging from long-medium term resources management (provisioning, resource pools) to flow level admission control, down to packet level scheduling and queuing management. Some of these mechanisms are related to NSs: in particular each NS is associated to a set of traffic handling algorithms at flow and packet level, collectively referred to as Traffic Classes (TCLs). This paper describes the set of traffic handling mechanisms defined in AQUILA, with a special focus on the implementation of TCLs, both at packet and flow level. In particular the scheduling/queuing and admission control schemes for each TCL are presented. Exemplary measurement results verifying the effectiveness of AQUILA approach for providing Quality of Service (QoS) guarantees and QoS differentiation are also included.

Virtualized Network Infrastructure Supporting Co-existence of Parallel Internets

In this paper we describe the IIP System designed, implemented and currently tested in the frame of the Polish national project entitled Future Internet Engineering1 (in Polish Inzynieria Internetu Przysz- IIP). The IIP system uses virtualized network infrastructure that allows to set a number of, named, Parallel Internets essentially differing in solutions for the data and control planes. More precisely, we design three Parallel Internets: (1) IPv6 QoS that basically explores Diff Serv and NGN architectures and uses TCP/IP, (2) Content Aware Network (CAN) specially designed for efficient content delivery, and (3) Data Streams Switching (DSS) for handling constant rate traffic with strong QoS requirements. For the IIP System we design a novel network infrastructure, which is based on the virtualization of the network elements (links and nodes). Such approach is on the line with the expectations of the network infrastructure for the Future Internet. In the paper we mainly focus on the system architecture, Parallel Internets and some aspects of network virtualization.

Admissible traffic load of real time class of service for inter-domain peers

The paper deals with the problem of assuring predefined QoS for real time (RT) class of service (CoS), which we define as one of inter-provider or inter-domain classes of service in IP network. We propose an analysis and adequate formulas to obtain admissible traffic load in a case when we map two end-to-end classes of service dedicated for telephony CoS and video conference CoS into one inter-domain RT CoS. The aim of the analysis is to determine the admissible load when the target packet loss ratio and buffer size dedicated for RT CoS are known. The proposed solution accounts for the difference between packet sizes of streams generated by voice (about 100 bytes) and video conference (rather 1500 bytes) applications. We illustrate our studies by simulation results

Designing the simulative evaluation of an architecture for supporting QoS on a large scale

The EuQoS system is a complete QoS system, scalable to large dimensions and addressing QoS at all relevant layers, which has been developed within the framework of the IST-EuQoS project. Its design has been aided by a considerable amount of modeling and simulation work, aimed at testing the various QoS mechanisms devised and their interaction. This paper describes the modeling and simulation work done in the framework of the project. We describe the three simulation models which have been developed, based on the different timescales at which the QoS mechanisms have effect. Furthermore, as a sample case of performance evaluation, involving different simulators, we describe the performance evaluation of the EuQoS signaling subsystem.

On the signaling system in the IPv6 QoS Parallel Internet

In this paper we provide an overview of the signaling system that we implemented in the IPv6 QoS Parallel Internet. This Parallel Internet has been proposed as one of Parallel Internets of the IIP System developed in scope of Future Internet Engineering project. The aim of the signaling system of the IPv6 QoS Parallel Internet is to perform call setup/release for assuring end-to-end QoS (Quality of Service) guarantees for selected data streams. For this purpose the proposed system follows current IETF and NGN (Next Generation Networks) recommendations. However, there are no detailed recommendations about the complete procedures in such signaling system of NGN architecture. Therefore, in this paper we propose the following procedures as exemplary scenarios: resource reservation call flow, resource termination call flow, notify/abort call flow. We also provide a sample usage of service stratum and transport stratum interface.

Admission control for TCP connections in QoS IP network

The paper describes a new admission control (AC) algorithm for greedy TCP connections. The algorithm has passed positive tests in pre-production QoS IP network [2], developed inside European IST project AQUILA network. The algorithm operates per single TCP flow and it assumes the setting of advertised receiver TCP window size for maintaining ideal TCP behaviour (no packet losses). The QoS objective is to guarantee the requested bit rate, say Rreq, a user demands from the network. Furthermore, on the basis of the Rreq and information about round trip time (RTT), the user request is mapped into the form of single token bucket parameters, i.e. token accumulating rate (R) and bucket size (BS), constituting input parameters for AC decision. For admitted TCP connection the Rreq is guaranteed, even if running connections differ in Rreq and RTTmin- Included simulation results confirm the efficiency of the algorithm.