Martin Karsten

An embedded charging approach for RSVP

Charging mechanisms are needed to protect an integrated services network from arbitrary resource reservations and to create a funding mechanism to extend network capacity at the most desired locations at the expense of those users that actually use these resources. In this paper, we describe a charging model that can be embedded in the RSVP (Resource ReSerVation Protocol) architecture. Our model is open and flexible in that it imposes little or no restrictions to the pricing policy of network providers or the usage behaviour of end-users. At the same time, it provides mechanisms to enable fine-grained charging of network communication. After a user-centric identification of requirements for charging mechanisms, a formal framework is presented to model the prices and payments. We present protocol elements and an implementation rationale to realize our charging model. Furthermore, we identify potential problems that are inherent to RSVP with regard to precise charging and we point out future research issues towards a realistic charging architecture.

Implementation and evaluation of the KOM RSVP engine

We describe implementation aspects and performance results of an innovative and publicly available RSVP implementation. Much debate exists about the applicability of RSVP as a signalling protocol in the Internet, particularly for a large number of unicast flows. While there has been a significant amount of work published on the theoretical concepts of RSVP signalling and conjectures about its presumed shortcomings, rather little attention has been paid to the implementation details of the core protocol engine. With our work, in spite of being still far from a final judgement, we try to shed light on this issue by presenting certain design details of a new implementation and a study about its performance. One particular result is given by the observation that a relatively cheap router based on PC hardware can sustain the signalling for more than 50,000 unicast flows.

Charging for packet-switched network communication-motivation and overview

Packet-switched network technology is expected to form the integration layer for future multi-service networks. The trend to integrate voice and data traffic can already be observed in the Internet and is expected to carry on until the full integration of all media types is achieved. At the same time it becomes obvious that the business model employed for current Internet usage is probably not sustainable for the creation of an infrastructure suitable to support a diverse and ever-increasing range of application services. Currently, the Internet provides only a single class of best-effort service and prices are mainly built on flat-fee, access-based schemes. However, the large and increasing differentiation of demand for quality of transmission services creates the necessity to differentiate service provision and, consequently, charging for network communications. In this article, we survey relevant service models and motivate the need for appropriate charging mechanisms. We give an overview on requirements for a charging system, its technical components, and issues for future business models that might by employed by service providers.

Aggregation of guaranteed service flows

It is common belief that the integrated services architecture (IntServ) is not scalable to large networks as, e.g., the global Internet. This is due to the ambitious goal of providing per-flow QoS and the resulting complexity of fine-grained traffic management. One solution to this problem is the aggregation of IntServ traffic flows in the core of the network. While one might suspect that aggregation leads to allocating more resources for the aggregated flow than for the sum of the separated flows if flow isolation is to be guaranteed, we show in this paper that for IntServs guaranteed service flows this is not necessarily the case even if flow isolation is retained. We compare different approaches to describing the aggregated traffic and analyze their impact on bandwidth consumption and ease of flow management. Applications of these theoretical insights could be to use the derived formulas for resource allocation in either a hierarchical RSVP/IntServ, IntServ over DiffServ (differentiated services), or IntServ over ATM network.

Design and Implementation of RSVP Based on Object-Relationships

RSVP has been proposed by the IETF as a signalling protocol for reservation-based quality-of-service enabled communication in IP networks. While RSVPs concepts are very sophisticated, further research efforts and potential modifications might be necessary to accomplish additional requirements before general deployment and commercial usage. Currently, only one freely available implementation exists and even some of the commercial implementations are based on it. In this paper, an alternative approach to describe RSVP protocol operations is presented, employing relational specification of state blocks and object-relationships between them. The result appears to be more concise and comprehensible than existing processing rules, yet not giving up efficiency. An implementation design based on this methodology, as well as specific details and optimizations are derived and explained. The implementation is designed to be portable across different operating system platforms and even to simulation environments. The primary purpose is to carry out research on modifications of RSVP, being able to examine those by simulation, emulation and real tests. Applying these considerations, an experimental protocol engine has been implemented, which is publicly available.

Cost and Price Calculation for Internet Integrated Services

Charging schemes are needed to protect an integrated services network from arbi- trary resource reservations and to create a funding mechanism to extend network capacity at the most desired locations at the expense of those users that actually use these resources. While not being the only input into pricing and charging, cost calculation is an important part of a charging scheme. In this paper, we develop a technique called virtual resource mapping to apply well- known economic principles to an optimal pricing framework and other tasks related to charging. Additionally, we describe how virtual resource parameters can be used to express prices when being combined with protocol mechanisms for charging. We focus on rate-based service guaran- tees in the context of Internet Integrated Services (IntServ) combined with IP multicast and RSVP as signalling protocol. It turns out that under given aggregated price-demand patterns, resource costs can precisely be extracted for each service request. Thereby, virtual resource parameters can be considered as link between economic theory and technical reality.

Experimental Extensions to RSVP - Remote Client and One-Pass Signalling

We present and evaluate two experimental extensions to RSVP in terms of protocol specification and implementation. These extensions are targeted at apparent shortcomings of RSVP to carry out lightweight signalling for end systems. Instead of specifying new protocols, our approach in principle aims at developing an integrated protocol suite, initially in the framework set by RSVP. This work is based on our experience on implementing and evaluating the basic RSVP specification. The extensions will be incorporated in the next public release of our open source software.

RSVP as firewall signalling protocol

Within a global networked environment, security aspects have become more and more important and access control at network borders is considered essential. For this purpose firewall systems are used which provide a well-established security mechanism to restrict the exchanged traffic to a certain subset of users and applications. In order to cope with the increasing demand for new applications, a firewall must be flexible and extensible to support such new applications and their protocols. RSVP is a dynamic signalling protocol, which has been invented to negotiate resource requirements between end systems and a packet-based communication network. We investigate the interoperation of RSVP with a firewall system in order to support new applications in a generic way. We show how the resulting system flexibility allows for a variety of employment scenarios and incremental deployment of such a technology. We back up our claims by describing a prototype that we have implemented.

Decoupling different time scales of network QoS systems

Providing quality of service (QoS) in large-scale networks like the Internet inherently needs to deal with heterogeneous network QoS systems. Therefore, the interworking between different network QoS systems is of high importance. In this paper, the interworking with respect to a basic characteristic of network QoS systems, the time scale of the system, is under investigation. The time scale of a network QoS system is its speed of reaction to individual requests for differentiated treatment of units of service. A slow time scale system will prefer requests to arrive with a low frequency and persist unaltered for a substantial period of time while a fast one is able to support much higher arrival rates of requests and is thus more amenable for short-lived units of service. Obviously, when overlaying a slow time scale QoS system over a faster one, there is no problem. However, and that is a more likely case, for the overlay of a fast time scale system on a slow one, there is a mismatch to be mediated at the edge between the two. The technique that is applied at an edge device for this mediation is called decoupling of time scales. Decoupling can also be viewed as aggregation of requests in time in contrast to spatial aggregation on the data path. In the paper, we develop an adaptive heuristic scheme to deal with decoupling and evaluate this scheme by extensive simulations.

Generalizing RSVP's traffic and policy control interface

In this paper we describe our current efforts to evaluate and extend the traffic control interface of RSVP and to define a useful policy control interface. The particular goal is to advance integration of non-broadcast multiple-access (NBMA) subnets, such as native ATM, and furthermore, to allow for meaningful inter-operation between traffic control and policy control. Our experience stems from implementing RSVP in combination with a policy interface to charging modules. Furthermore, we have developed a VC management module for ATM, which enables flexible inter-operation of RSVP with ATM subnets. We describe the current status of this work with respect to multicast, atomicity of operations and resulting challenges.