An adaptive load-aware burst assembly scheme to achieve optimal performance of FDL buffers in OBS networks

Optical burst switching (OBS) is regarded as one of the most promising switching technologies for next generation optical networks. Contention resolution of data bursts is a critical mission to implement practical OBS. The use of fiber delay line (FDL) buffers has received a lot of attention as a fundamental but effective solution to resolve burst contention. Several studies have investigated the way to achieve the optimal performance of FDL buffers at a single-node level. However, this article studies how to achieve the best performance of OBS networks with FDL buffers under varying traffic condition at a network level. For this purpose, we propose an adaptive load-aware burst assembly (ALBA) scheme, which adaptively adjusts the size threshold of burst assembler optimized to the current network traffic load. A piggybacking method used to deliver the traffic-load information from core nodes to ingress edge nodes accelerates the adaptiveness of the proposed scheme by reducing the update time of the size threshold. The effectiveness of the ALBA scheme is proved by comparing with No-FDL case and fixed size-threshold cases under changing traffic-load environment from extensive simulation tests.

Physical impairment aware scheduling in optical burst switched networks

Optical burst switching (OBS) presents itself as a promising technology for bridging the gap between optical wavelength switching and optical packet switching. Increasingly, researchers attempt to incorporate more realistic constraints into the design of OBS networks. Optical signal transmission quality is subject to various types of physical impairment introduced by optical fibers, switching equipment, or other network components. The signal degradation due to physical impairments may be significant enough such that the bit-error rate of received signals is unacceptably high at the destination, rendering the signal not usable. In this paper, based on earlier work, we study the burst scheduling problem in OBS networks, taking into account physical impairment effects. We propose three effective burst scheduling algorithms: (1) a JET based Physical Impairment Constrained Algorithm (JETPIC), (2) an Integrated Physical Impairment Constrained Algorithm (IPIC), and (3) an Enhanced Integrated Physical Impairment Constrained Algorithm (EIPIC). At an OBS node, the proposed algorithms schedule bursts for transmission by searching for available resources as well as verifying signal quality. Our simulation results show that the proposed algorithms are effective in terms of reducing the burst blocking probability. In general, algorithm JETPIC outperforms algorithms IPIC and EIPIC in burst blocking probability and average end-to-end delay performance.

Long-term estimation-based burst control algorithm in OBS networks

By taking advantage of statistical multiplexing gain in the burst level, optical burst switching (OBS) technology enables optical Internet to handle huge volume of data in an efficient manner without requiring optical buffers in the optical domain. However, when congestion builds up in the optical network core, large amount of data might be lost. In this article, we propose an efficient optical burst control algorithm that operates based on the awareness of future burst traffic condition to eliminate the effect of congestion reaction delay. The proposed algorithm takes advantage of multiple statistics to improve the estimation accuracy.Through performance evaluation, it is verified that the proposed algorithm proactively controls inbound burst traffic so that the OBS network can stay in a stable traffic condition while keeping the network throughput high.

Congestion window-based adaptive burst assembly for TCP traffic in OBS networks

Burst assembly is one of the key factors affecting the TCP performance in optical burst switching (OBS) networks. When the TCP congestion window is small, the fixed-delay burst assembler waits unnecessarily long, which increases the end-to-end delay and thus decreases the TCP goodput. On the other hand, when the TCP congestion window becomes larger, the fixed-delay burst assembler may unnecessarily generate a large number of small-sized bursts, which increases the overhead and decreases the correlation gain, resulting in a reduction in the TCP goodput. In this paper, we propose adaptive burst assembly algorithms that use the congestion window sizes of TCP flows. Using simulations, we show that the usage of the congestion window size in the burst assembly algorithm significantly improves the TCP goodput (by up to 38.4% on the average and by up to 173.89% for individual flows) compared with the timer-based assembly, even when the timer-based assembler uses the optimum assembly period. It is shown through simulations that even when estimated values of the congestion window size, that are obtained via passive measurements, are used, TCP goodput improvements are still close to the results obtained by using exact values of the congestion window.     

On burst-header contention-resolution in OBS networks

Advances in enabling technologies and the explosive growth of Internet traffic has led to the widespread proliferation of network systems in recent years. With their relatively low cost, high throughputs, high-bandwidth utilization, and low-transmission latency, Optical Burst Switching (OBS) networks represent an ideal solution for next-generation Internet applications. However, in OBS networks, Burst Header Packet (BHP) contentions occur when two or more BHPs are switched simultaneously to the same output port of a given core node. These contention events result in significant losses of the corresponding data burst. Accordingly, this study presents a Store-and-forward COntention-REsolution mechanism, designated as SCORE, which utilizes fiber delay line buffers to resolve the BHP collision problem, thereby minimizing the burst loss rate. The results of a series of simulations performed using an OIRC OBS-ns simulator confirm the effectiveness and efficiency of the proposed scheme.

A new fault-management method using congestion-avoidance routing for optical burst-switched networks

Optical burst switching (OBS) has been proposed as a promising technique to support high-bandwidth, bursty data traffic in the next-generation optical Internet. This study investigates a new fault-management framework for an OBS network. In an OBS network, burst-loss performance is a critical concern. In OBS, the data-burst transmission is delayed by an offset time (relative to its burst control packet (BCP), or header), and the burst follows its header without waiting for an acknowledgment for resource reservation. Thus, a burst may be lost at an intermediate node due to contention, which is generally resolved according to the local routing and bandwidth information. The routing table maintained in each OBS node is generally pre-computed and fixed to forward the data bursts. Such a static forwarding feature might have limited efficiency to resolve contentions. Moreover, a burst may be lost and the network may be congested when a network element (e.g., fiber link) fails. Therefore, for reliable burst transport, we develop dynamic routing approaches for preplanned congestion avoidance. Our goal is to proactively avoid congestion during the route-computation process, and our approach employs wavelength-channel utilization, traffic distribution, and link-distance information in the proposed objective functions for routing. Two update mechanisms for maintaining a dynamic routing table are presented to accommodate bursty data traffic. Based on our routing mechanisms, we also propose a new congestion-avoidance-and-protection (CAP) approach, which employs a primary route and a group of backup routes for each node pair against failures and congestion. The performance of the proposed protection strategy using congestion-avoidance routing is demonstrated to be promising through illustrative numerical examples.

A new proportional differentiation scheme based on batch scheduling for optical burst switching networks

Optical burst switching (OBS) is a promising switching paradigm for building the next generation optical internet. The proportional differentiation model is very convenient for network operators to quantitatively adjust the quality differentiation among service classes. To provide proportional differentiated services for OBS networks, a proportional differentiation scheme based on batch scheduling is proposed in this article. The scheme adopts the batch scheduling idea to reserve data channel resources for a batch of data bursts. It helps to decrease burst dropping probability. When some data bursts are unsuccessfully scheduled, a preemption method is used to keep a proportional burst dropping probability among service classes according to the expected burst dropping probability equations given by the proportional differentiation model. The scheme has low computational complexity. Simulation results show that the scheme can provide proportional differentiated services and efficiently decrease the burst dropping probability.

TCP over OBS: Impact of consecutive multiple packet losses and improvements

In TCP over OBS networks, consecutive multiple packet losses are common since an optical burst usually contains a number of consecutive packets from the same TCP sender. In this paper, we first present a new theoretical method to analyze the behavior of Reno when consecutive multiple packet losses occur. Results of the analysis indicate that even a small number of consecutive multiple packet losses can force Reno to timeout. Then we propose B-Reno, a newly designed TCP implementation that can overcome Reno’s inefficiency in dealing with consecutive multiple packet losses over OBS networks and can avoid the shortcomings of New-Reno and SACK. Results of comprehensive simulations indicate that B-Reno over OBS networks can achieve a performance better than Reno and New-Reno, and that it can also achieve a performance similar to that of SACK. Moreover, B-Reno only needs some simple modifications to New-Reno at the sender’s protocol stack, and thus has less difficulty in deployment and less protocol complexity than that of SACK.

Improving throughput of long-hop TCP connections in IP over OPS networks

Optical Packet Switching (OPS) can provide the ever-increasing bandwidth required for Internet traffic and new applications for future networks. However, optical packet loss is the major problem for an OPS network. Moreover, by increasing the number of hops between a pair of ingress–egress switches in an OPS network, optical Packet Loss Rate (PLR) between this pair is increased. Therefore, due to a higher PLR for long-hop TCP connections, the throughput of these connections may be much lower than the short-hop TCP connections. To overcome this problem, it is proposed in this paper to use the retransmission idea in the optical domain not only to increase TCP throughput but also to improve the throughput of multi-hop TCP connections, and also to have a loss-free OPS network. Under retransmission in the optical domain, a copy of the transmitted traffic is kept in the electronic buffers of ingress switches and retransmitted in the optical domain whenever required. Note that the TCP layer has its own retransmission at the client packet level as well. By retransmission of lost packets in the optical domain, TCP would be unaware of the lost client packets, and therefore, TCP would not reduce its sending rate. In this paper, TCP throughput is studied in a bufferless slotted OPS network and the effectiveness of the proposed mechanism is evaluated.

nOBS: an ns2 based simulation tool for performance evaluation of TCP traffic in OBS networks

Performance evaluation of tcp traffic in obs networks has been under intensive study, since tcp constitutes the majority of Internet traffic. As a reliable and publicly available simulator, ns2 has been widely used for studying tcp/ip networks; however ns2 lacks many of the components for simulating optical burst switching networks. In this paper, an ns2 based obs simulation tool (nobs), which is built for studying burst assembly, scheduling and contention resolution algorithms in obs networks is presented. The node and link objects in obs are extended in nobs for developing optical nodes and optical links. The ingress, core and egress node functionalities are combined into a common optical node architecture, which comprises agents responsible for burstification, routing and scheduling. The effects of burstification parameters, e.g., burstification timeout, burst size and number of burstification buffers per egress node, on tcp performance are investigated using nobs for different tcp versions and different network topologies.     

A frequent-pattern approach for optical networks routing planning

Optical burst switching (OBS) networks have been receiving much attention as a promising approach to build the next generation optical Internet. In the bufferless DWDM switching technology, burst loss that should be minimized is the key design parameter. One of the critical design issues in OBS network is how to plan the optimal routing path in order to minimize burst dropping due to network resource contention. This study proposes the burst frequent-pattern tree (BFP-Tree) approach to pre-determine a suitable routing path in the OBS network. The BFP-Tree approach essentially is a learning-based mechanism that is able to determine a suitable transmission path from the historical network transaction data. The experiment results show that the successful rates of routing paths obtained by the BFP-Tree approach are able to converge to those of the optimal results.

Assembly admission control based on random packet selection at border nodes in Optical Burst-Switched networks

In Optical Burst-Switched (OBS) networks employing Just Enough Time (JET) signalling, Burst-Control Packets request resources at intermediate nodes following a one-way reservation protocol, that is, requested resources are not confirmed back to the source. Hence, data bursts are transmitted without any guarantees, and it sometimes occurs that these are dropped at a certain hop in the source–destination path, hence wasting resources at previous hops. This effect is specially harmful if some connections are abusing of the global shared resources, violating their respective Service Level Agreements, thus causing: (1) global performance degradation; and, (2) unfair service received by other connections. This article proposes “Random Packet Assembly Admission Control”, an admission control mechanism for OBS networks that moderates the two problems above. The mechanism monitors the network load status, detects which links are heavily loaded and decides which flows among the total traversing them require throughput decrease, on attempts to alleviate congestion and benefit other flows which are not abusing from the network. Such throughput decrease consists of preventive packet dropping during the assembly process at the ingress nodes of the OBS network, thus making no use of the network core. The numerical results show a substantial increase in the throughput experienced by well-behaved flows, and fundamental fairness achievement in the use of optical resources.

Design and performance comparison of multiple-token based MAC protocols for optical burst switched ring networks

Optical burst switching (OBS) has been proposed as a new optical switching paradigm for the next generation Internet due to its flexibility and feasibility compared to OCS and OPS. Moreover, serving as a backbone that interconnects a number of access networks, OBS ring topologies have been a good choice for solving the current metro gap problem between core network and access network owning to its simplicity and scalability. In this paper, we provide an insight into the OBS ring network that consists of nodes using TT–TR (Tunable Transmitter–Tunable Receiver). The node architectures with TT–TR may make efficient use of network resources even though traffic pattern, such as IP traffic with self-similarity dynamically change, and can support good expandability. However, all nodes share the limited network resources. This may result in contention such as wavelength contention and transceiver contention leading to burst loss. In order to use the shared network resources fairly and efficiently as well as reducing the resource contention, we focus on the design of medium access control (MAC) protocols based on multiple tokens. Each token is allocated to one wavelength to denote the accessibility of that wavelength, i.e., once the token is captured, the corresponding wavelength can be used to transmit a burst. As tokens hold the key for using wavelengths to transmit bursts, token management including the token release time is crucial in the proposed MAC protocols. Thus, two kinds of multiple-token based MAC protocols with different token release times are proposed: token release after transmitting burst (TRTB) and token release after transmitting control header (TRTC). Each of them is classified into two schemes called TRTB/TRR and TRTB/RCA and correspondingly TRTC/TRR and TRTC/ RCA. RCA stands for receive collision avoidance. The target is to increase the performance while reducing the processing overhead at each node. The performance of the TRTB and TRTC protocols are evaluated and compared in terms of queuing delay, burst loss rate, and channel utilization by OPNET simulation. The effects of various design parameters are also investigated through simulation in order to evaluate their scalability. In all the proposed schemes, tokens are just used to denote the accessibility of each wavelength. Finally, as an alternative, we also propose a new scheme based on the TRTC protocol called TRTC/CAT (collision avoidance by tokens) to avoid contention by using tokens.

在OBS网络中改进TCP性能的高效机制

Transmission Control Protocol (TCP) performance over Optical Burst Switching (OBS) is experimentally investigated on an OBS network testbed, concluding that burst losses will lead to a significant drop in the available TCP bandwidth. Two mechanisms are introduced to improve TCP performance. One is concerning the burst assembly optimization and another is based on the novel assembly and scheduling mechanism to reduce the burst losses.     

On the fairness improvement of channel scheduling in optical burst-switched networks

In the past years, several signaling protocols were proposed for OBS networks and the most popular one is the Just-Enough-Time (JET) protocol. JET not only efficiently utilizes the network capacity, but also effectively reduces the end-to-end transmission delay. However, the most critical defect of JET is its intrinsic deficiency: Fairness. The fairness problem is a traditional problem common to various kinds of networks. It results in a phenomenon that bursts with a shorter number of hops are generally favorized and hence deteriorates the network utilization as well. In this article, we investigate this problem and propose a fair channel scheduling algorithm as a solution. Usually there is a tradeoff between fairness and blocking performance. Accordingly, the objective of our scheme is to achieve a balance between the two conflicting metrics as much as possible. In our scheme, each burst is associated with a dynamic priority which is defined by several characteristics of the burst. When contention occurs, the proposed scheme picks the preferable burst and drops the other one according to their priorities. From simulation results, we observed that the proposed scheme could improve fairness without causing significant reduction in dropping performance. Furthermore, it increases the effective link utilization as well.

Amplification effects of the send rate of TCP connection through an optical burst switching network

This paper investigates support for TCP RENO flows in an Optical Burst Switching (OBS) network. In particular we evaluate the TCP send-rate, i.e., the amount of data sent per time unit taking into account the burst assembly mechanism at the edge nodes of the OBS network and burst loss events inside the network. The analysis demonstrates an interesting phenomenon, that we call correlation benefit. This phenomenon is introduced by the aggregation mechanism and can give rise, in some conditions, to a significant increase in the TCP send-rate. These results are obtained by means of an analytical model, based on a Markovian approach, and have been validated via an intensive simulation campaign.     

Burst-cluster transmission with probabilistic pre-emption for reliable data transfer in high-performance OBS networks

Recently, optical switching and packet processing technologies have been developed and high-performance optical burst switching (OBS) networks are constructed by using these technologies. In high-performance OBS networks, several types of applications such as Grid computing and HDTV can be provided for users according to immediate reservation protocol. Because some applications require that data is transmitted reliably over high-performance OBS networks, it is indispensable to provide reliable data transfer service for high-priority users. Therefore, in this paper, we propose a reliable burst transmission method which can be available for the immediate reservation protocol. In the proposed method, both burst-cluster transmission and probabilistic pre-emption are used at edge and core nodes. By using these methods together, the reliable data transfer and the service differentiation can be provided. We evaluate by simulation the performance of the proposed method in the 14-node NSFNET. Numerical examples show that the proposed method can transmit higher-priority bursts more reliably than the conventional method while not increasing the overall burst loss probability so much. In addition, we investigate effective parameter settings from some simulation results.

Feasibility analysis for service differentiation using an FDL bank in OBS networks

A service differentiation scheme in optical burst switching (OBS) networks, which is based on dynamic fiber delay line (FDL) assignment, is shown. The effectiveness of the scheme is validated by numerical analysis and extensive simulations. Especially, the feasibility conditions for the service differentiation scheme, which are considered as the minimum number of FDLs for each sub-FDL group, are displayed. The feasibility conditions are derived numerically, and are verified through extensive simulations. The results of extensive simulations show that the proposed scheme and the feasibility conditions are valid for service differentiation in OBS networks.

An Optimization Framework for Demand-based Fair Stream Allocation in MIMO Ad Hoc Networks

In this paper, we investigate the problem of scheduling flows for fair stream allocation (or, stream scheduling) in ad hoc networks in which the transmitter and receiver use multiple antennas called Multiple Input Multiple Output (MIMO) technology. Our main contributions include: i) the concept of stream allocation to flows based on their traffic demands or class, ii) stream allocation to flows in the network utilizing single user or multiuser MIMO communication, iii) achieving the proportional fairness of the stream allocation in the minimum possible schedule length, and iv) performance comparison of the stream scheduling in the network for single user and multiuser communication and the tradeoff involved therein. We first formulate demand-based fair stream allocation as an integer linear programming (ILP) problem whose solution is a schedule that is guaranteed to be contention-free. We then solve this ILP in conjunction with binary search to find a minimum length contention-free schedule that achieves the fairness goals. We show that an implementation of our algorithm for a number of sample topologies in fact yields minimum length schedules that achieve the fairness goals. We then give performance comparison results that show the benefit of multiuser MIMO links over single user links at higher traffic workloads in the network. Finally, we also give a greedy heuristic for stream scheduling and compare its performance with the ILP-based algorithm in terms of the fairness goals achieved in a given schedule length.

Framework for Analysis of Transmission Rate Scheduling for Wireless CDMA Data Networks

The optimal utilization of network resources and the capacity to fulfill quality of service requirements are key requirements for 3rd G networks operations. Several burst admission and transmission rate scheduling algorithms are proposed in the literature. In this study, we develop an analytical framework for the downlink transmission rate scheduling problem for CDMA networks employing discrete service bit rates. The framework uses the average downlink transmit power as the system state and develops a K-dimensional Markov chain representing all possible states in the system. In addition, the transition probabilities due to arrivals of burst requests are made a function of the power utilization and the average power required to support the new burst taking into account the path loss model. The study assumes a transmission rate assignment scheme where the maximum possible system bit rate is assigned given the current system state and the subscriber’s eligibility profile for particular system service rates. The analytical model provides performance metrics such as system throughput, average power utilization, average number of simultaneous transmissions, burst request blocking probability, and mean burst service time. While the developed model is applicable for a general CDMA based network with arbitrary discrete system service bit rates, the model is evaluated for the example of a cdma2000 1xRTT network. A comparison between simulation and analytic results to assess the accuracy of the model is provided.