While being critical to the network management, the current state of the art in network measurement is inadequate, providing surprisingly little visibility into detailed network behaviors and often requiring high level of manual intervention to operate. Such a practice becomes increasingly ineffective as the networks grow both in size and complexity. In this paper, we propose vPROM, a vSwitch enhanced SDN programmable measurement framework that automates the measurement process, minimizes the measurement resource usage, and addresses several significant technical challenges faced by early works. vPROM leverages the SDN programmability and extends the Pyretic runtime system and OpenFlow network interface to achieve the measurement automation. The required measurement resources are minimized by only acquiring the necessary statistics, made possible with instrumented Open vSwitches1with user defined monitoring capability. By decoupling monitoring from routing, vPROM reduces the interference between the measurement applications and other applications, and eliminates the frequent involvement of the controller. A vPROM prototype is implemented with DDoS and port-scan detection applications. The performance of vPROM is evaluated and the comparison results with other existing programmable measurement approaches are also presented.

In this paper, we present a discussion of some architectural principles pertaining to mapping distribution protocols that are common components of proposed designs to improve the scalability of future Internet routing tables. The efficiency of a mapping distribution protocol in terms of response time and the volume of traffic load it generates are important considerations. We consider how Egress Tunnel Routers (ETRs) can perform aggregation of end-point ID (EID) address space belonging to their downstream delivery networks, in spite of migration/re-homing of some subprefixes to other ETRs. This aggregation may be useful for reducing the processing load and memory consumption associated with mapping messages, especially in resource-constrained components of the mapping distribution system. Consequently, the proposed methods can also help improve response time (e.g., first packet delay). Some key architectural issues, their potential solutions and trade-offs are discussed. The overarching goal is to expose and discuss some subtleties in design considerations for mapping distribution and management.

In this paper, a Multi-hop B3G architecture, which is a Heterogeneous Wireless Network (HWN) architecture, integrating MANET, Infrastructure WLAN with 3G UMTS network, is considered for the purpose of providing anytime, anywhere data access as it utilizes the higher bandwidth rates of MANET and WLAN networks and wider range of communication of 3G network. The issues of Gateways in HWN, which serve as a liaison between the MANET/WLAN and UMTS, are analyzed in this paper. A multi-metric Gateway Selection mechanism using Simple Additive Weighting (SAW) of MANET node metrics such as residual energy, 3G signal strength and mobility speed, is used to select an optimal Gateway from the dual-interface-enabled Gateway Candidates. As transaction progresses, when the above-mentioned metrics of the Gateway reach their respective pre-defined threshold values, thereby, affecting optimality of the current Gateway, a Multi-metric Adaptive Gateway Migration Mechanism (MAGMM) is proposed, which migrates the responsibility of the current Gateway to a newly-selected optimal Gateway, in order to sustain connectivity of the integrated network. This MAGMM is incorporated in the considered multi-hop B3G architecture and hence it is termed as Adaptive Gateway Management-based Multi-hop B3G Network (AGMMB3G) Architecture. In addition to MAGMM, this paper also compares AGMMB3G with the other existing HWN architectures. Simulations are carried out in HWN environment using NS2 and the results show how MAGMM enhances the overall HWN performance in terms of Transaction Duration, Packet Delivery Ratio and Control Packet Overhead parameters.

There are currently heterogeneous wireless access networks. A large variety of applications utilizing these networks will demand features such as real-time, high-availability and even instantaneous high-bandwidth in some cases. Therefore, it is imperative for network service providers to make the best possible use of the combined resources of available heterogeneous networks for connection support. Thus, with regard to vertical handoff performance, there is a critical need for developing algorithms for connection management and optimal resource allocation for seamless mobility. In this paper, we develop a vertical handoff decision algorithm that enables a wireless access network to not only balance the overall load among all attachment points but also to maximize the collective battery lifetime of mobile nodes (MNs). Simulation results are also presented to demonstrate the efficacy of the proposed algorithms.

We present a detailed study of the potential impact of border gateway protocol peering session attacks and the resulting exploitation of route flap damping (RFD) that cause network-wide routing disruptions. We consider canonical grid as well as down-sampled realistic autonomous system (AS) topologies and address the impact of various typical service provider routing policies. Our modeling focuses on three dimensions of routing performance sensitivity: 1) protocol aware attacks (e.g., tuned to RFD); 2) route selection policy; and 3) attack-region topology. Analytical results provide insights into the nature of the problem and potential impact of the attacks. Detailed packet-level simulation results complement the analytical models and provide many additional insights into specific protocol interactions and timing issues. Finally, we quantify the potential effect of the BGP graceful restart mechanism as a partial mitigation of the BGP vulnerability to peering session attacks.

We propose the creation of a wireless storage area network (SAN) and analyze its benefits. The proposed wireless SAN (WSAN) consists of a SAN switch that is connected to multiple wireless access points (APs) that communicate with the storage devices. This network would save space and reduce overall costs by not requiring wired connections. Wireless SANS would also provide more freedom in the placement of storage devices. However, because the number of wireless access points is less than the number of storage devices, it is possible for user data requests to be blocked if all access points are busy. An important design goal is therefore to minimize the probability that a network access request will be blocked.

Agile all optical cross-connect (OXC) switches currently use an architecture in which regenerators and transceivers have pre-assigned and fixed directionality. However, technology is evolving to enable new OXC architectures in which the directionality of regenerators and transceivers can be dynamically assigned on demand for each connection that requires regeneration. We have performed detailed analytical and simulation studies to compare the two architectures. The analytical study is applicable to a single node and is very useful in providing intuitive insights into the two alternative architectures. The simulation study is based on a realistic network topology consisting of 53 nodes. The simulation study was carried out using NIST's updated GMPLS lightwave agile switching simulator (GLASS) tool. The GLASS tool was significantly enhanced over its previous version in the course of this study. We report extensive results on comparison of the two OXC architectures in a realistic network implementation in terms of connection blocking probability, efficiency of regenerator use, and carrier equipment costs. We show that fewer regenerators and transceivers need to be used with the new architecture because of sharing of resources across all directionality combinations. This translates to significant cost savings for the new architecture, especially as the traffic load in the network increases.

Optical burst switching (OBS) is a very promising switching technology for realization of an economical optical Internet. In OBS networks, when contention occurs at an intermediate switch, two or more bursts that are in contention can be lost because a forwarding path reservation is not made for a burst until a control message for the burst arrives. That is the reason why one of the critical design issues in OBS is finding ways to minimize burst dropping resulting from resource contention. In this paper, we propose and analyze a novel deflection routing protocol, which mitigates and resolves contention with significantly better performance as compared with techniques currently known in the literature. While several variants of the basic deflection routing scheme have been proposed before, they all lacked the ability to determine the alternate route based on clear performance objectives. In this paper, we present an on-demand deflection routing scheme, which sequentially performs the following: 1) based on certain performance criteria, dynamically determines if the burst should be deflection routed or retransmitted from source and 2) if the decision is to deflection route, then the same is done using a path that is based on minimization of a performance measure that combines distance and blocking due to contention. The proposed contention-based limited deflection routing scheme prevents injudicious deflection routing. Our simulation results show that the scheme proposed here has much superior performance both in terms of burst loss probability and increased network throughput. Through analytical and simulation modeling, a number of useful insights into the OBS network protocols and performance are provided.

The introduction of new high bandwidth services such as video-on-demand by cable operators will put a strain on existing resources. It is important for cable operators to know how many resources to commit to the network to satisfy customer demands. In this paper, we develop models of voice and video traffic to determine the effect on demand growth on hybrid fiber-coax networks. We obtain a set of guidelines that network operators can use to build out their networks in response to increased demand. We begin with one type of traffic and generalize to an arbitrary number of high-bandwidth CBR-like services to obtain service blocking probabilities. These computations help us to determine how cable networks would function under various conditions (i.e., low, medium, and heavy loads). We also consider how the growth rate of the popularity of such services would change over time, and how this impacts network planning. Our findings will help cable operators estimate how much bandwidth they need to provision for a given traffic growth model and connection blocking requirement.

As network utilization continues to grow in the coming years, there will be increased pressure on network operators to use traffic engineering to provision resources more efficiently. One way to do this is to allow backup paths associated with disjoint working paths to share bandwidth. Increasing the amount of sharing will naturally increase the risk that a failed working path will either be unrecovered or forced to use dynamic recovery mechanisms. To examine the tradeoffs between robustness and efficiency and to develop useful performance bounds, we develop theoretical models for (1:1)nrecovery schemes that are independent of the network's topology and management plane. We confirm our results using simulations of uncorrelated failures in a wide-area optical network with various degrees of resource sharing.

One of the critical design issues in optical burst switching (OBS) networks is finding ways to minimize burst dropping resulting from resource contention. The existing variants of the basic deflection routing schemes all lack the ability to determine the alternate route based on clear performance objectives. In this paper, we present contention-based limited deflection routing (CLDR) scheme, which sequentially performs the following: (1) based on certain performance criteria, dynamically determines if the burst should be deflection routed or retransmitted from source, (2) if the decision is to deflection route, then the same is done using a path that is based on minimization of a performance measure that combines distance and blocking due to contention. Through analytical and simulation modeling, a number of useful insights into the OBS network protocols and performance are provided.

This paper investigates the packet delay variation and packet loss probabilities to be expected in typical applications of voice trunking using ATM adaptation layer type 2 (AAL2) and the consequences these may have, such as the need for sequence numbers in voice packet headers. Simulations have been run for the case of 32-kb/s ADPCM encoded voice across a physical link whose bandwidth is 1/4, 1/2, or 1 times DS1 (1.536 Mb/s). The results show the number of calls that can be admitted while meeting the ninety-ninth percentile queuing delay objective of 20 ms or less. The delay budget that can be allocated to packet queuing delay at access is about 20 ms, given a one-way end-to-end delay budget of 100 to 150 ms for conversational voice. The number of calls is also limited so that the ratio of late or lost packets does not exceed 10/sup -3/. Voice quality is known to be tolerant to packet losses of up to one in 100. When the packet delay variation is significant, as it threatens to be in this situation, sequence numbers can help a receiver to detect and to recover from anomalies-lost, early, or late packets-and make fewer reconstruction errors than if sequence numbers were absent or ignored. Three circumstances are identified where packet slips can occur with ripple effects. An analysis of these cases leads to lower bounds on the sequence number modulus. The appropriate value of the sequence number modulus depends on the desired call load and performance objectives, and can be deduced from the results presented.

Asynchronous transfer mode (ATM) adaptation layer 2 (AAL2) has been designed for efficient transport of voice, fax, and voiceband data (VBD) traffic over an ATM virtual circuit. The protocol helps achieve low latency and high bandwidth efficiency while applying suitable compression methods on voice/VBD/fax calls and silence elimination on voice calls. We analyze the performance and capacity of an ATM multiplexer based on AAL2 adaptation. We assume that embedded adaptive differential pulse code modulation (ADPCM) is used to compress voice, and silence elimination is used to achieve statistical multiplexing gain. The embedded ADPCM coding scheme allows selective dropping of less significant bits of voice during congestion in the ATM/AAL2 multiplexer. We compare the call capacities of voice multiplexers with and without bit dropping (BD). The performance models and results presented are based on fairly general assumptions and can be used for traffic engineering and call admission control in land-line or wireless ATM systems for a variety of voice/voiceband compression algorithms. A generalized algorithm for call admission control is also described.

Some methods of supporting voice in broadband ISDN, (B-ISDN) asynchronous transfer mode (ATM), including voice compression, are examined. Techniques for voice compression with variable-length packet format at DS1 transmission rate, e.g., wideband packet technology (WPT), have been successfully implemented utilizing embedded adaptive differential pulse code modulation (ADPCM) coding, digital speech interpolation (DSI), and block-dropping schemes. For supporting voice in B-ISDN, voice compression techniques are considered that are similar to those used in WPT but with different packetization and congestion control methods designed for the fixed-length ATM protocol at high speeds. Possible approaches for packetization and implementation of variable-bit-rate voice coding schemes are described. ADPCM and DSI for voice coding and compression and cell discarding (CD) for congestion control are considered. The advantages of voice compression and CD in broadband ATM networks are demonstrated in terms of transmission bandwidth savings and resiliency of the network during congestion.>

An analytical model is presented for computing the end-to-end voice call performance in a packet network which drops the less significant bits in voice packets during periods of congestion. The authors first modify an existing single-node bit-dropping model to include the situation when the overall arrival process at a node consists of a mix of packets of different sizes. Variability in packet size is caused by bit dropping at previous nodes. A detailed model is presented to capture bit dropping effects in a tandem connection of nodes. The model includes the effects of load fluctuations at each node, and also takes into account the dependencies in bit dropping experienced by a voice packet at successive nodes in a tandem connection. These dependencies occur because a packet that has experienced bit dropping at a node in congestion state Q would not be further bit-dropped at a subsequent node if its current congestion state is no worse than Q. The model also incorporates the reduction in packet service times at intermediate nodes due to dropping at previous nodes. In addition to the detailed model, the authors present two approximation procedures which also serve as upper and lower bounds. In particular, the upper bound is shown to be very tight for practical range of loads, and hence serves as a good approximation with significant computational simplicity.>

Analytical models are presented for computing the end-to-end voice call performance in a packet network that drops the less significant bits in voice packets during periods of congestion. These models provide information about the end-to-end quality likely to be experienced in future packet-switched integrated services networks. An existing single-node bit-dropping model is modified to include the situation resulting when the overall arrival process at an internal node consists of a mix of packets of different sizes due to bit dropping at previous nodes. A detailed model to capture bit-dropping effects in a tandem connection of nodes is presented. The model includes the effect of load fluctuations at each node, and also takes into account the dependencies in bit dropping experienced by a voice packet at successive nodes in a tandem connection. The model also incorporates the internodal dependence when reductions in packet service times occur at intermediate nodes due to bit dropping at previous nodes. Two approximation procedures are discussed that serve as upper and lower bounds. In particular, the upper bound is shown to be very tight for a practical range of loads, and hence serves as a good approximation with significant computational simplicity.>

The performance of a packet voice multiplexer queue in which the less significant bits of voiced packets are dropped during states of congestion in the multiplexer is examined. Using the results of simulation and analytical modeling, it is illustrated that bit dropping of voice packets significantly smooth the burstiness of superposition packet voice traffic by speeding up the packet service rate during critical periods of congestion in the queue. The smoothing effect renders it possible to approximate the superposition by a Poisson process for modeling a packet voice multiplexer with bit dropping. By comparison with a simulation, an analytical model based on the Poisson assumption is shown to produce quite accurate performance predictions. The results indicate that significant capacity and performance advantages are gained in the multiplexer as a result of the bit-dropping scheme.>

The authors examine the performance of a packet voice multiplexer queue in which the less significant bits on voice packets are dropped during states of congestion in the multiplexer. Using the results of simulation and analytical modeling, it is illustrated that bit dropping on voice packets significantly smoothes the superposition packet voice traffic by speeding up the packet service rate during critical periods of congestion in the queue. This phenomenon renders it possible to approximate the superposition process by a Poisson process for analyzing a packet voice multiplexer with bit dropping. The multiplexer is modeled using an M/D/1/K model in which D denotes the deterministic but state-dependent nature of service. By comparison with a simulation, this model is shown to produce quite accurate performance predictions.>

This paper analyzes a model of a multiplexer for packetized voice and data. A major part of the analysis is devoted to characterizing the aggregate packet arrival process resulting from the superposition of separate voice streams. This is done via the index of dispersion for intervals (IDI), which describes the cumulative covariance among successive interarrival times. The IDI seems very promising as a measurement tool to characterize complex arrival processes. This paper also describes the delays experienced by voice and data packets in the multiplexer using relatively simple two-parameter approximations.

Discrete-time analysis of two schemes for multiplexing voice and data is presented. In each scheme voice and data are multiplexed using the movable boundary frame allocation scheme. In the first scheme, speech activity detectors (SAD's) are not used, and hence, the variations in the voice traffic are only due to the on/off characteristics of voice. In the second scheme, SAD's are employed so that talker silences can he utilized for transmission of additional voice and/or data. In this scheme, the multiplexer performs digital speech interpolation as well as movable boundary frame allocation. The performance measures considered are probability of loss for voice calls, probability of speech clipping, speech packet rejection ratio, and the expected data message delay. In the case of the multiplexer with SAD, a tradeoff exists between data message delay and speech interpolation advantage. Some numerical examples are presented which illustrate the performance of the two multiplexers.