The main architectural problem with the current inter-domain routing model is the overloading of IP addresses due to their use for identification, location, and forwarding. This makes it hard to support the need for multi-homing, traffic engineering, and mobility while maintaining address hierarchy that is essential for scalable routing. Thus, there is a need to separate identifiers from locators. In this paper we identify architectural and protocol changes in addressing and routing needed to provide major improvements in scalability. We design a corresponding protocol solution (Address-Identifier Mapping System — AIMS) based on separation of identification and addressing, and a mechanism for mapping ID to address.

Voice and video traffic can tolerate some errors in end-to-end transfer but require very low latency. Data traffic typically requires an almost error free transfer but can tolerate higher latency. Thus, coding and forward error corrections are the main mechanisms for error controls for voice and video traffic while ARQ is the main mechanism for error controls for data traffic. When the error rates are high (for example, in wireless links), it is beneficial to have both forward error corrections and ARQ for data traffic. In addition, the effectiveness of the coding and FEC can be improved by using interleaving. The combined effects of coding, interleaving, and ARQ on data traffic effective throughput is not well understood. In this paper, we take the first steps. In particular, we show the effective throughput is very sensitive to channel, coding, and interleaving parameters, thus identifying a need to study parameter optimization further. We also show that appropriately sized LDPC codes provide better throughput than traditional codes do. Early results also suggest that, for mixed traffic environment, it may pay to have some coding and interleaving be done at higher layer to allow application specific parameters.

The simple data link (SDL) is a new framing protocol for variable/fixed length packets over a general-purpose point-to-point communications channel. SDL extends the HEC-like framing mechanism used in ATM to variable-length data. Its low implementation complexity makes it particularly suitable for high-speed transport links in wavelength channels and dark fiber applications. SDL is a very attractive alternative to conventional solutions such as ATM and PPP-over-SONET (POS) for high data rate environments. SDL has also been designed to facilitate many OAM&P functions needed in next generation multiservice optical packet networks including multiprotocol encapsulation, virtual links, quality-of-service (QoS) differentiation, link management, and control. We describe framing and data link synchronization procedures in SDL, and evaluate its performance over octet and bit synchronous transport facilities.

Integrated Network Design Tools (INDT) is a suite of tools being developed at Bell Labs. This suite of tools is built on a flexible software architecture based on C++ and can provide optimal network design for a mix of private line, switched voice and switched multimedia services using a mix of circuit and ATM switches and cross connects and facilities with different bandwidth granularity. INDT also has the ability to design, in a standalone mode or in conjunction with logical designs, the core transport network using mesh and/or ring topologies and using PDH or SONET/SDH products. The transport network may be designed over existing fiber routes and capacities or the tool may be used to help decide the fiber layout. We focus on a new capability within INDT, viz., SONET ring design. We present a high level description of the various submodules in the ring network design and provide an overview of the main factors that influenced our algorithm development. We also present sample studies of our approach to intra-ring load balancing as well as our economic quantification of penalties due to dual ring interconnect (DRI) and other features of interest.

Asynchronous transfer mode (ATM) is now well recognized as the fundamental switching and multiplexing technique for future broadband ISDN. As these networks will be increasingly relied upon for providing a multitude of integrated voice, data, and video services, network reliability is a key concern. There are several intrinsic features of ATM networks that could potentially be exploited to provide improved restoration techniques, beyond those established for synchronous transfer mode (STM) networks, such as digital cross-connect restoration or self-healing rings. These features include ATM cell level error detection, inherent rate adaptation and nonhierarchical multiplexing. The authors explore the use of these features in developing fast restoration strategies for ATM networks. In particular, they address: (1) ATM error detection capabilities for enhanced failure detection, (2) network rerouting strategies, (3) spare capacity allocation, and (4) network control architecture and related implementation aspects. Their findings suggest that fast network span failure detection and bandwidth-efficient rerouting capabilities can be combined to develop restoration strategies for ATM networks with significantly greater performance-cost ratios when compared to existing STM network restoration strategies.>

The cellular radio link is characterized by deep fades leading to long error bursts (lasting hundreds of milliseconds). Data transmission over such links results in large packet error rates (in the range 10/sup -3/ to 10/sup -1/). We have designed a radio link protocol (RLP) to achieve high throughput on such links. The RLP is based on frequent, complete or partial feedback of the receiver state. Performance results for the US digital cellular TDMA standard show that the scheme can provide the equivalent of 9.6 kb/s service per full-rate TDMA channel above a carrier to interference ratio of 18 dB.>

The performance of the SNR protocol of A. N. Netravali et al. (1990) is studied when it is implemented for end-to-end flow and error control. Using a combination of analysis and simulation, the efficiency with which this protocol uses the network bandwidth and its achievable throughput is evaluated as a function of certain network and protocol parameters. The protocol is enhanced by introducing two windows to decouple the two functions of receiver flow control and network congestion control. This enhancement and the original protocol are compared with go-back-N (GBN) and one-at-a-time-selective-repeat (OSR) retransmission procedures, are shown to have significantly higher throughput for a wide range of network conditions. As an example, for a virtual circuit with 60-ms roundtrip delay and 10/sup -8/ bit error rate, in order to deliver 500 Mb/s throughput, both the GBN and OSR require a raw transmission bandwidth of approximately 800 Mb/s, whereas SNR with two windows needs slightly higher than 500 Mb/s raw bandwidth. Periodic exchange of state can also provide a variety of measures for congestion control in a timely and accurate fashion.>

The author discusses two in-call buffer/window negotiation schemes for short intermittent file transfers, focusing on the specific needs of such traffic streams. A closed network of queues models is developed to show the behavior of the proposed schemes. These models, while simple, capture essential details of the control schemes. Under fairly general assumptions, the resulting network of queues is of product form and can be analyzed using the mean value analysis. The author uses such an analysis to compare the proposed schemes and to determine appropriate sizes of trunk buffers to achieve the desired balance between bandwidth utilization and file transfer delay. The effects of other parameters on the performance of these schemes as well as on the buffer sizing rules are discussed. In-call parameter negotiation may be carried out by the end system with the network elements or by the access controller with the internal network elements. The author discusses implications of this location as well as the needed protocol features. The service discrimination capabilities desired at the trunk controller in switching modes are discussed.>

The authors evaluate the performance of both the fixed buffer allocation (FBA) and the adaptive buffer allocation (ABA) schemes, in which the network nodes are allowed to offer less than the requested buffer size. The performance measures of interest are the blocking probability, file transfer delay, and the adaptation speed for ABA for a given buffer size and the offered load. The authors develop and analyze a quasi-birth-death model of the ABA scheme (with exponential file lengths and negligible delay in carrying out reservation and cancellation procedures). In particular, they develop a recursive computational scheme exploiting the structure of the underlying model. This is supplemented by a first-passage time analysis to evaluate the transient behavior of the control strategy. The authors use both analytic and simulation methods. The results demonstrate that the ABA schemes provide significant advantages over the FBA scheme if the parameters are appropriately chosen. They also provide guidelines on the choice of these parameters.>

In a Variety of overloaded queueing systems (e.g., an overloaded call processing system), long delays can result either in poor service given to the customer or in customers, unknown to the system, turning "bad." For example, in switching systems, long dial tone delays can result in customers initiating dialing before receiving dial tone. In this case the system will not receive all the digits and an unsuccessful call results. This can lead to the system expending real time on unsuccessful services and, therefore, reduces the effective throughput. Thus, there is a need for control Schemes which reduce the load offered to the processor by selectively refusing service to some customers in such a way as to keep delays, for those customers which are selected for service, small. This fact has been recognized and has led to improved strategies for local switches. In this paper we analyze and compare the performance of various queueing and service disciplines for anM/M/1queue. We consider LIFO and FIFO Schemes with customer rejection mechanisms corresponding to pushing out or timing out older customers in queue. Delay distributions for served customers are obtained and comparisons based upon throughput-delay tradeoffs are presented. For the situation where Customers can turn "bad" at a random time after their arrival, we compare the throughput of good customers. The results presented are a mixture of classical results, which are briefly stated, and new results which are developed in more detail. The numerical results show a dramatic effect of the queueing and Service disciplines on the overload performance and a strong dependence of the throughput of successful services on the mechanism for customers turning "bad." Although results are obtained for a single server queue, they can be used to approximately analyze overload control schemes which control access to distributed systems.