光突发交换网络基于优先级与突发包分割的光缓存方法

为了保证光突发交换(OBS)网络中不同优先级业务的服务质量(Qos)和有效地降低突发包的丢失率,提出了一种基于优先级与突发包分割的光缓存方法.该方法通过在输出数据信道上设置光纤延迟线组来缓存被分割的冲突突发包.当冲突发生时,基于突发包的优先级进行"竞争突发包头部分割或者原突发包尾部分割,,处理;无冲突部分进行交换或者直...     

光突发交换网络中多优先级业务冲突性能分析

对在光突发交换网络核心节点中发生的多优先级业务冲突问题进行了性能分析,与以往同类研究不同的是,不仅给出了不同优先级业务在核心节点产生的阻塞率情况,而且还给出了各优先级业务平均进入系统的数据量以及占用系统的均值等性能,研究结果对设定网络流量以及业务优先级等问题提供了理论参考价值.     

OBS网络中突发重传和IP包重传的性能比较

文章分别分析了将基于优先级的突发重传和基于分段的IP包重传方式引入光突发交换(OBS)网络后的性能,并对这两个机制进行了分析比较.通过分析比较发现,采用基于优先级的突发重传可以有效地减少重传次数、所需缓存和重传延迟,提高网络吞吐量,是OBS网络中的一种理想的重传选择方案.     

OBS网络基于光缓存的多优先级业务冲突研究

光突发交换(OBS)网络是下一代IP over WDM光网络的典型代表.丈章时配置有光缓存的OBS网络核心节点中发生的多优先级业务冲突问题进行了性能分析,给出了不同优先级数据包的丢失率和平均等待时延,研究结果可以为OBS网络的实用化提供理论参考.     

OBS网络中基于优先级的先分割后缓存冲突解决方法

为了有效地降低突发包的丢失率和保证OBS网络中不同优先级业务的服务质量,提出了一种基于优先级的先分割后缓存冲突解决方法。该方法通过在多波长信道系统的输出数据信道上设置光纤延迟线组来缓存被分割的冲突突发包。当冲突发生时,首先基于突发包的优先级和处理完毕时间进行"竞争突发包头部分割或者原突发包尾部分割"处理;无冲突部分进行交换或者直接在事先预留的输出数据信道上处理,冲突部分的分割突发包进行光缓存。仿真结果表明,多波长信道系统中,高优先级突发包的丢失率低于低优先级突发包的丢失率。同时该方法在一定程度上可以有效地减少端到端的传输时延和整个网络的丢包率,从而提高整个OBS网络的性能。     

OBS网络中基于优先级抢占的恢复机制

提出一种新的支持光突发交换网络中优先级抢占的快速恢复机制.该机制对不同优先级的业务采取不同的恢复策略,实现故障的快速恢复.同时采用阈值检测方式支持优先级抢占,减少受影响链路中较高优先级业务的丢包.仿真结果表明,基于优先级抢占的恢复机制具有较低的丢包率,对高优先级业务丢包率性能改善更为明显.     

一种能有效支持QoS的光突发装配技术

提出了一种新的光突发装配技术，目的在于增强光突发交换(OBS)网络的服务质量(QoS)能力。该技术由自适应门限突发装配算法和基于优先级的随机化的偏移时间设置方案构成。该装配算法特别适合于多类别分组混合装配，能让所有类别的分组公平有效地使用装配能力，能较好地与IP层的QoS机制相匹配。偏移设置方案将突发控制分组(BCP)与突发数据(BP)间的偏移时间分成QoS偏移和随机偏移，前者由改进的JET协议按照有选择性的突发段丢弃机制(BSSD)确定，后者则由令牌桶机制确定。BSSD仅丢弃包含有低类别分组的突发段，而非整个突发。计算机仿真结果表明，本文提出的突发装配技术在性能上具有优越性。     

OBS网络中一种基于突发包优先级分割的可控重传方案

在光突发交换(OBS)网络中,突发包会由于竞争OBS核心节点输出端口的有限波长资源而发生冲突。突发包重传能够在一定程度上减少由于突发包在核心节点冲突而导致的数据损失,但重传次数的增加可能会加重网络负荷,反而增加数据丢失率。并且,在多业务存在的OBS网络中,重传方案需要能够实现区分服务以保证网络的服务质量(QoS)。据此,本文提出一种基于突发包优先级分割的可控重传方案,在实施优先级分割的同时,根据网络负荷赋予每次重传不同的概率,并对重传次数加以控制。最后,仿真分析了路径阻塞率和不同优先级业务的字节丢失率(ByLP,byte loss probability)性能。     

OBS基于回退和信道分级的冲突解决算法

解决突发包的冲突问题是光突发交换网络的关键技术之一。为了降低突发包冲突时丢包率,提出了一种新型的基于信道分级和回退机制的冲突解决算法,该算法考虑突发业务优先级,将信道分成两个级别,有区别地对待高低优先级两种业务类型,并且增加了一部分回退信道作为保护和重发冲突数据包,能够有效地保证高优先级业务的传输可靠性。结果表明,当冲突发生时,网络能根据此时高低优先级业务量的比重配比,自适应地调节两个级别信道的数量,兼顾到低优先级业务的低丢包率,保证了较低的延时率。此算法相比以往的冲突解决算法具有较低的丢包率和相对小的延时率。     

OBS网突发包竞争问题的解决方案

光突发交换(OBS)是近几年出现的一种很有前途的新型光交换技术。由突发包之问竞争所导致的数据丢失是光突发交换网络所面临的一个重要问题，它影响到光突发交换能否走向实用化。为此提出了一种用于解决突发包之间竞争钧综合方案，并对它的性能进行了仿真研究。仿真结果表明，该方案可以大幅度地降低OBS网络中的数据丢失事。     

Burst scheduling for differentiated services in optical burst switching WDM networks

Optical burst switching (OBS) is one of the most important switching technologies for future optical wavelength division multiplexing (WDM) networks and the Internet. The model of differentiated services has been proposed to support quality of service (QoS) in the IP‐based Internet. It is also very important to have differentiated service support in OBS networks. When the burst scheduling in an OBS network is set up appropriately, network can support differentiated services. In this paper, we proposed a new burst scheduling scheme, called differentiated scheduling with identical priority offset time (DSIPO). In DSIPO, the same priority offset time is used for all the bursts destined to the same edge node regardless of their priorities. Differentiated services in terms of burst loss probability are achieved by processing the control packets of higher priority class bursts, thus reserving resources for their data bursts, more promptly upon their arrival than those of lower priority class bursts. Each intermediate (core) node can adjust the burst loss probabilities of various burst classes by choosing its own differentiated processing delay value for each priority class or its own differentiated processing delay difference value between any pair of adjacent priority classes. We model and analyse DSIPO in terms of the burst loss probability for each priority class with simulation validation. The performance of DISPO is evaluated by simulation. Copyright © 2004 John Wiley & Sons, Ltd.     

Dynamic wavelength selection and delayed burst assignment schemes for Optical Burst Switching networks

Optical Burst Switching (OBS) has been proposed as a promising switching technology for the next generation of optical transport networks. In this paper, we address the issue of how to provide proportional differentiated services in OBS networks. Firstly, a Dynamic Wavelength Selection (DWS) scheme is introduced to provide proportional differentiated services in bufferless OBS networks by dynamically assigning more and longer periods of wavelengths to high priority classes. This scheme can also utilize wavelengths efficiently because the wavelengths are shared among different classes. Next, a Delayed Burst Assignment (DBA) scheme is introduced, by which bursts of the high priority class are given a higher probability for reserving wavelengths by scheduling the bursts of the low priority class with a delay to provide quality of service (QoS) in OBS networks. The integration of these two schemes provides proportional differentiated services and improves the burst loss performance by giving the burst head packet (BHP) two opportunities of scheduling its data burst (DB).     

Hybrid offset‐time and burst assembly algorithm (H‐OTBA) for delay sensitive applications over optical burst switching networks

Optical burst switching (OBS) is the most favourable switching paradigm for future all‐optical networks. Burst assembly is the first process in OBS and it consists of aggregating clients packets into bursts. Assembled bursts wait for an offset time before being transmitted to their intended destinations. Offset time is used to allow burst control packet reserve required resources prior to burst arrival. Burst assembly process and offset‐time create extra delay in OBS networks. To make OBS suitable for real time applications, this extra latency needs to be controlled. This paper proposes and evaluates a novel offset time and burst assembly scheme to address this issue. Constant bit rate (CBR) traffic that has stringent end‐to‐end delay QoS requirements is used in this study. The proposed scheme is called hybrid offset‐time and burst assembly algorithm (H‐OTBA). The objective of the paper is achieved by controlling maximum burst transfer delay parameters of CBR. The proposed scheme was evaluated via network simulation. Simulation results demonstrate that, H‐OTBA has effectively reduced end‐to‐end delay for CBR traffic compared with current solutions. Copyright © 2014 John Wiley & Sons, Ltd.     

Rate-controlled optical burst switching for both congestion avoidance and service differentiation

Optical Burst Switching (OBS) has recently been proposed as a candidate architecture for the next generation optical Internet. Several challenging issues remain to be solved to pave the way for the OBS vision. Contention arises in OBS networks when two or more bursts are destined for the same wavelength, and a wide variety of reactive contention resolution mechanisms have been proposed in the literature. One challenging issue in OBS is proactively controlling the traffic flowing through the OBS network so that the network does not stay in a persistent state of contention, which we call the congestion avoidance problem. Another challenging issue is the need for service differentiation, which is common today in electronically switched networks via the use of advanced buffer management and scheduling mechanisms. However, such mechanisms cannot be used in OBS networks due to the limited use, or total absence, of buffering. One of the popular existing approaches to service differentiation in OBS networks is the use of larger offset times for high-priority bursts which, however, increases the delays and may adversely affect application-level performance. In this paper, we propose a feedback-based rate control protocol for the control plane of the OBS network to both address the congestion avoidance and service differentiation issues. Using this protocol, the incoming traffic is dynamically shaped at the edge of the OBS network in order to avoid potential congestion in the burst-switched core. Moreover, the traffic shaping policies for the low and high priority traffic classes are different, and it is possible using the proposed protocol to isolate high-priority and low-priority traffic almost perfectly over time scales on the order of a few round-trip times. Simulation results are reported to validate the congestion avoidance and service differentiation capabilities of the proposed architecture.     

一种新型的OBS网络单向资源预留方法

光突发交换(OBS)是实现IP/DWDM网络的一种极有前途的方案。该文着重讨论了光空发交换(OBS)网络中单向资源预留的方式。文中首先介绍了著名的JET(Just Enough Time),提出了一种新的单项资源预留方式 Just Little Time (JLT),它不但能使用偏转路由,还能有效利用Fiter Detag Line（FDL）延时来弥补偏移时间的不足,只需使用很小的初始偏移时间,降低了数据在网络中的延时。随后,文章讨论了使用偏移时间增量来实现区分服务的方法。计算机仿真结果证明了所提方法的有效性。     

Providing service differentiation for optical-burst-switched networks

This paper proposes a service differentiation scheme called preemptive wavelength reservation protocol (PWRP) to provide proportional quality of service (QoS) for optical-burst-switched (OBS) networks. In the context of service differentiation, traffic is divided into different service classes based on traffic characteristics. A service differentiation scheme then provides different degrees of resource assurance to different classes of traffic in proportion to their service classes. Unlike existing approaches, which may degrade to classless schemes or which may suffer from low wavelength utilization, the mechanism in this paper is robust and efficient and supports an incremental deployment of QoS support. A usage profile for each class is maintained at the router, and a preemptive wavelength reservation mechanism is implemented to ensure QoS. An analytical model is derived and simulations are conducted to evaluate the performance. The results show that the approach described in this paper performs better than existing mechanisms in terms of lower blocking probability and higher resource utilization, making it an excellent QoS mechanism for OBS networks.     

Group-Scheduling for Multi-Service Optical Burst Switching (OBS) Networks

We propose a new burst scheduling mechanism for Optical Burst-Switched (OBS) networks. The new approach is made possible by gathering data bursts into groups and performing the scheduling decision for each group collectively. In OBS group-scheduling, bursts will not be considered for scheduling until a pre-defined time period elapses, during which the group of burst header packets would be gathered. By transforming a set of data bursts into a set of corresponding time intervals, the problem of scheduling these bursts is transformed into a combinatorial optimization problem. Graph algorithms are applied to obtain the maximum number of non-overlapping bursts. The proposed OBS group-scheduling scheme is shown to improve the performance of OBS networks over existing scheduling schemes in terms of burst loss probability and channel utilization. With an extension through a sequential optimization, using a Branch-and-Bound technique, the proposed scheme can support multiple classes of service. It is shown that the new scheduling approach has several desired characteristics including fairness and service differentiability among classes in terms of burst loss probability and channel utilization.     

一种OBS网络绝对区分服务实现机制

该文提出了一种用于OBS网络的绝对区分服务实现机制PSPJET.该机制结合使用额外偏移时间和Burst抢占/Burst主动分段丢弃技术,既能确保保证业务的数据丢失率不高于要求的最大丢失率,还能有效降低非保证业务的数据丢失率,提高网络性能.和现有的APJET机制相比,PSPJET机制对高业务强度的承受能力更好,还具有无需使用光缓存的优点.     

Exact Analysis of Offset-Based Service Differentiation in Single-Channel Multi-Class OBS

We study a multi-class optical burst switching (OBS) node using the horizon reservation scheme. Multiple traffic classes are differentiated using different offset times per class. Assuming Poisson burst arrivals and phase-type distributed burst lengths, we exactly solve for per-class blocking probabilities using the well-known theory of feedback Markov fluid queues.     

A new composite assembly mechanism for supporting QoS in OBS networks

To provide the differential quality of service（QoS）for different classes of packets and reduce the packet loss probabil ity（PLP）,a novel priority-based composite assembly scheme for optical burst switching（OBS）networks is proposed The low and high packet classes are aggregated into a single burst simultaneously,and the highest-priority packets ar placed in the middle,while the low-priority packets are at the tail and head of the burst.The priority is lowered gradu ally from the middle to the ends.Simulation results demonstrate that the proposed assembly strategy not only guaran tees the integrity of the high-priority bursts,but also significantly reduces the average end-to-end delay of the burst and the PLP of network.So it can adapt to the flexible network with QoS requirement.