Peter Pieda

Bandwidth assurance issues for TCP flows in a differentiated services network

Much industry attention has been focused on providing differentiated levels of service to users on IP networks. One such proposal is the RIO scheme proposed by Clark (see ACM Transactions on Networking, 1998 ). RIO is an extension of the RED algorithm that relies on a differentiated drop treatment during congestion to cause different levels of service. The end result of differentiated dropping of packets during congestion is differentiated throughput rates for end-users. The IETFs Diffserv Working Group has recently standardized a PHB (per hop behaviour) that is based on a differentiated drop scheme-assured forwarding (AF). This paper raises issues with providing bandwidth assurance for TCP flows in a RIO-enabled differentiated services network. The main contribution is a detailed experimental study of five different factors that impact throughput assurances for TCP and UDP flows in such a network. Our study demonstrates that these factors can cause different throughput rates for end-users in spite of having contracted identical service agreements.

Intelligent Traffic Conditioners for Assured Forwarding Based Differentiated Services Networks

Issues related to bandwidth assurance in Assured Forwarding based Differentiated Services (Diffserv) networks have been discussed in recent research papers [7][8][11]. Some of the factors that can bias bandwidth assurance are Round Trip Time (RTT), UDP/TCP interaction and different target rates. The bias due to these factors needs to be mitigated before bandwidth assurance for a paying customer can be articulated in Service Level Agreements (SLAs). This paper proposes intelligent traffic conditioning approaches at the edge of the network to mitigate the effect of Round Trip Time, UDP/TCP interactions, and different target rates. The simulation results show a significant improvement in bandwidth assurance with intelligent traffic conditioning. The limitation of the proposed solutions is that they require communication between edge devices. In addition, these solutions are not applicable for a one-to-any network topology.

Using TCP models to understand bandwidth assurance in a Differentiated Services network

In this paper, a comprehensive analytical model to predict the bandwidth achieved by aggregates of TCP flows in a DiffServ network is presented. The model predicts achieved bandwidth in three different cases: an over-provisioned network, an under-provisioned network, and a near-provisioned network. In developing the model, we ensure that all parameters are measurable using standard tools and information available from routers and network management tools in todays networks. Simulation was used to establish the validity of the model and understand its scope of applicability and limitations. Using the model, we explain why achieved excess bandwidth is based on factors such as RTT, packet size, and CIR. Finally, we present a novel extension of the model to predict the bandwidth of TCP flows in a Diffserv network with multiple congested nodes.

Experimental study of assured services in a diffserv IP QoS network

Much attention has recently been given to the differentiated services (Diffserv) approach to provide Quality of Service (QoS) for IP networks. This packet-marking based approach to IP QoS is attractive due to its simplicity and ability to scale. Two of the most popular services proposed for the Diffserv approach are the Assured and Premium Services. In this work prototypical implementations of Diffserv components are described. The prototypes are used to study the single-queue, dual drop-preference model proposed as a basis for assured services in Diffserv.

Improving Network Infrastructure Security using Geospatial Technology

Advanced network management systems (NMS) have made tremendous strides in terms of visualization, monitoring, and vulnerability analysis for Layer 2 and Layer 3 data networks. However, they are unable to take into account geographic considerations and vulnerability of the network to physical threats and hazards. This paper summarizes our recent work on utilizing geospatial mapping and technology for improving network infrastructure security. We propose a novel architecture and requirements for integrating NMS and GIS (geographic information systems) technology to address issues such as the one mentioned above. A prototype system was developed to validate the architecture. We conclude the paper with a discussion of challenges encountered during development and testing of the prototype on an emulation test-bed