Conference PaperPDF Available

Contention resolution for optical burst switching networks using alternative routing

Authors:

Abstract

Optical burst switching (OBS) is an approach to building very high capacity routing switches based on optical data paths and electronic control. Therefore, OBS is considered as a viable solution in the high-speed optical Internet backbone. In this paper, we propose an intra-class contention resolution scheme in optical burst switching networks by the enhanced alternative routing algorithm. The blocking probability of the proposed scheme is significantly reduced as compared with the general OBS while the additional delay due to the alternative path is smaller than 10% of the basic offset time.
1
NGI seminar 1
Contention Resolution for Optical
Burst Switching Networks Using
Alternative Routing
(Mar. 11, 2003)
Yonsei Univ.
Computer Communication Lab.
Kim Hyo Jin
NGI seminar 2
Contents
Abstract & INTRODUCTION
OPTICAL BURST SWITCHING
ALTERNATIVE ROUTING IN OBS
NETWORKS
SIMULATION AND RESULTS
CONCLUSION
2
NGI seminar 3
Abstract & INTRODUCTION
Dense Wavelength Division Multiplexing(DWDM)
technology
9Exploit the huge potential bandwidth
9Without any opto-electro-optic(O/E/O) conversions.
Optical Burst Switching(OBS)
9To build very high capacity routing switches.
9Good solution in high-speed optical Internet
backbone.
9Eliminate the electronic bottleneck
9at switching node without O/E/O conversion
9Guarantee the Class of Service(CoS)
9without any buffering.
NGI seminar 4
Propose :
Propose :
9
9Alternative routing based intra
Alternative routing based intra-
-contention
contention
resolution scheme by suggesting a control
resolution scheme by suggesting a control
header packet format
header packet format
9The blocking probability is significantly reduced.
Additional delay
for the alternative path Basic offset time< 10% x
3
NGI seminar 5
OPTICAL BURST SWITCHING
OBS composition
9OBS network
9Optical core routers and electronic edge routers
9Each burst
9Control header packet and a data burst
9Control header contains
9Information of the data burst length and offset time
Packets are
9Assembled into bursts at the network ingress.
9Disassembled back into packets at the network
egress.
NGI seminar 6
Two coupled overlay networks
9A pure optical network transferring data bursts
and a hybrid control network transferring control
header packets.
9Control network
9A packet switched network
9Controls the routing and scheduling of data bursts in the
all optical network
9based on the control header packets information
9Advantages
9Mature electronic control technologies
9Promising optical transport technologies
4
NGI seminar 7
Fig. 1 : three burst switching techniques
9A burst can cut through intermediate routers.
NGI seminar 8
(a) In-Band Terminator(IBT) based
burst switching
9Bandwidth is reserved
9As soon as IBT
9Followed end of the data burst
9Limitation
9Be processed all optically
9Not support optical processing
5
NGI seminar 9
(b) Tell-And-Go(TAG) based burst switching
9Operating steps
Source node sends a control packet over the control
channel to reserve a bandwidth.
Corresponding data burst is transmitted without waiting
for the acknowledgement.
The source node sends a release packet to the
destination.
9Ex) Just-In-Time(JIT) protocol
9Disadvantage
9Large signaling overhead
NGI seminar 10
(c) Reserve-Fixed-Duration(RFD) based
burst switching
9Bandwidth is reserved
9fixed time duration at each router
9Information of reservation time
9Carried by control packet
9Ex) Just-Enough-Time(JET) protocol
9Eliminates the signaling overhead
9Improve the resource utilization
9Disadvantage
9Complexity increased.
6
NGI seminar 11
Many challenges to implement the
OBS network
9Edge router
9Burst offset time management, burstification
and burst assembly mechanism
9Core router
9Data burst and control header packet
scheduling, protection and restoration
mechanism and contention resolution scheme
NGI seminar 12
We focused
We focused
9Contention resolution
9The core router has no fiber delay line(FDL) and the
offset time is fixed at the ingress router.
9Classify the contention in intermediate core
routers.
9The inter-class contention caused by different class
bursts.
9The intra-class contention caused by same class bursts.
9To prevent the inter-class contention
9Additional QoS offset time can be used and high priority
bursts can avoid blocking caused by low priority bursts.
9No published result yet.
7
NGI seminar 13
ALTERNATIVE ROUTING IN
OBS NETWORKS
Our three methods for solving the
intra-contention problem in core
routers
The wavelength converter
FDL
The deflection routing
We proposed..
9
9An alternative routing scheme
An alternative routing scheme
9
9Combines wavelength converters and the enhanced
Combines wavelength converters and the enhanced
deflection routing algorithm.
deflection routing algorithm.
NGI seminar 14
If contention occurs at intermediate
core routers, the data burst is blocked.
9Because it cant change its path.
In our algorithm
In our algorithm
9
9Every core router has hop by hop routing
Every core router has hop by hop routing
functions.
functions.
9
9The routing table has two different paths
The routing table has two different paths
for each destination.
for each destination.
8
NGI seminar 15
NGI seminar 16
Operating steps
9If contention occurs at certain intermediate core router,
the hop by hop routing functions are performed.
9One is routed through the short path and the other is routed
through the alternative path obeying certain policy(time
critical burst, number of hops, etc.)
9If next node does not encounter any contention, the
rerouted data burst will be routed through the shortest
path.
9The control packets contain the number of rerouting times
9Thus if the number of rerouting is over the threshold value,
the rerouted bursts are just drop to prevent looping and early
arrival of data bursts.
Better than deflection routing
9
NGI seminar 17
Routing offset time
9The additional offset time
9to make up the time by alternative routing.
9Depends on
9the network size and the header processing
time
9Ingress edge routers decide the offset time
9Prevents the data burst surpasses the
control header.
NGI seminar 18
10
NGI seminar 19
Message type
9The header packet type
9Uniquely determines the exact layout of the message.
QoS
9Following data bursts priority.
Input port
and
Input wavelength identifier
9Indicate the incoming data bursts port and wavelength.
Offset time
9The separation time between control header and data burst
9Consists of the basic offset time and routing offset time.
Burst length
9Length of data burst
Hop count
9Prevent bursts not to lose their destination router or the looping
environment.
Source address and Destination address
9Addresses of the source and the destination.
NGI seminar 20
SIMULATION AND RESULTS
Java language
9To evaluate the performance of blocking probability
9in the slotted 4x4 Manhattan Street network.
In fig. 2
9Assumed
9Nodes
90,3,12,15 nodes edge routers
9Others core routers
9The bursts arrive synchronously.
9Each time slot is equal to the burst duration.
9Available wavelength at each node is one.
9For simplify the simulation
9Input traffic
9Exponential distribution and Pareto distribution with Hurst parameter
H=0.9.
11
NGI seminar 21
Fig. 5 and Fig. 6
9Blocking probability is reduced by 10 and 2 times
9When offered load is 0.1 and 0.9.
9Self-similar traffic with H=0.9,
9The proposed alternative routing scheme performs very well.
NGI seminar 22
Fig. 7
9Alternative Path makes the additional delay(additional hop
counts) performance.
12
NGI seminar 23
CONCLUSION
An intra-contention resolution
scheme
9By introducing additional routing offset
time in bufferless intermediate core
routers in the OBS network.
9Using Routing offset time
9Effective to significantly reduce the blocking
probability for the exponential
... Regardless of the limitations of switching speed of the switching technology (OPS or OBS), our attention moves towards the buffering part. So far, FDL has been used for optical buffering [2][3][4]. A microsecond of delay requires about 200 m of fiber. ...
Article
Full-text available
The discrete time delays are essential for resolving contention for the same wavelengths between arriving bursts. The Fiber Delay Lines (FDLs) are used to provide a delay since optical random access memory is unavailable. The time period which is common between two bursts is known as contention period and if it is larger then the burst is blocked. As a result, a limited contention period resolution capacity is available. This limitation leads to burst dropping and extended burst delay. In this paper, we have proposed a buffer architecture to improve the management of contended bursts. The contented bursts are handled with priority order depending upon their size. We have proposed the processing of bursts with calculated delay using FDLs by sending every contented burst through FDLs that provide sufficient delay.
... Output port contention can occur due to unavailability of a wavelength at the desired output port for an incoming burst. Several techniques exist in the literature to avoid contention such as FDLs[43,44], Deflection Routing[45,46], Wavelength Conversion[47] and Segmentation Based Dropping[48,49] but none of them can guarantee zero burst loss. OBS with a two-way reservation scheme ensures zero burst loss in which a control packet reserves resources across all intermediate nodes from the source to the destination and is sent back to the source as an acknowledgement. ...
... Researcher proposed an intra-class contention resolution scheme in optical burst switching networks by the enhanced alternative routing algorithm. The blocking probability of the proposed scheme is significantly reduced as compared with the general OBS while the additional delay due to the alternative path is smaller than 10% of the basic offset time [1]. The first optical switching paradigm that has been proposed in the literature is the Optical Burst Switching with burst drop. ...
Article
Full-text available
Optical burst switching is a promising solution for terabit transmission of IP data bursts over WDM networks. Optical Burst Switching (OBS) is a promising paradigm for the next-generation Internet. In this paper, we propose a random algorithm, first-fit wavelength algorithm and wavelength reservation algorithm. The blocking probability of the Optical Burst Switch will be analyzed depending upon the number of channels, loads and the number of links. The number of wavelengths on all the links will be kept constant. The blocking probability of the proposed scheme is significantly reduced as compare with general OBS.
Article
Full-text available
With the propelling high capacity demands, passive optical networks (PONs) are getting extra consideration nowadays. Problem: Burst blocking in wavelength division multiplexed (WDM) PON system increase with the contention period > burst time. Motivation: In the absence of optical random memory in optical communication networks, contention resolution using temporal delays is prominent work and can be accomplished using Fiber temporal delay paths. Burst blocking becomes high in the network at larger contention period such as time between two bursts is larger. Methods: Therefore, in this work, a 40 Gbps × 10 WDM channels based PON system with 50 GHz channel spacing over 20 km is investigated by incorporating switches, switch controller, buffers and Fiber temporal delay paths to decrease the bursts blocking in the system. In order to assign delay to bursts, an algorithm is proposed and calculated amount of delay are assigned to demanding wavelengths for bursts which inturn decrease the blocking. Analysis: Proposed WDM-PON system is simulated in the simulation tool and investigated for different FTD delays, burst loss probability, burst arrival rate, buffering delay and performance is also analyzed in terms of log BER. Results: Bursts blocking probability increases with the increase in burst arrival rate, lesser buffering delays, and more no. of input ports. Further investigation reveals that higher extinction ratios can provide better log BER.
Article
The most significant difficulties in optical burst switching networks is the nature of service arrangement. A few propositions have been made and a few restrictions have been watched taking into account the material used and existing contention resolution techniques in optical networks. An intelligent effective strategy that improves the quality of service in such networks dependent on intelligent water drops(IWDs). Movement of burst depends upon the material used in optical fibre and scheduling algorithms used for scheduling of bursts. Intelligent water drops strategy manages optical burst misfortune during conflict goal measure and offers a separated support procedure utilizing organized dispute goal arrangements in the network centre. Likewise, wasteful information channel arrangement calculation element burst misfortune favored OBS. The proposed strategy is researched through broad simulations. The simulation results show that using intelligent water drops burst loss rate can be decreased for both low and high traffic loads contrasted with existing schemes.
Chapter
OBS networks using the JET/JIT reservation mechanism, which is a one-way and unacknowledged reservation mechanism, provide connectionless switching. Due to this feature, the bursts may contend with one another at the core nodes. Contention among two bursts occurs due to the overlap of two or more bursts (in time) that arrive simultaneously on two different links or wavelengths and request the same wavelength at a given time. In electronic packet switching networks, contention is handled by buffering. However, in optical networks, buffers are difficult to implement and there is no optical equivalent of random access memory. When multiple bursts contend for a wavelength at an intermediate node, all but one of the contending bursts is dropped. Burst loss due to contention is a major source of concern in OBS networks. Since these losses are quite random in nature and occur due to lack of buffers, it poses a serious problem in the deployment of OBS technology in networks that support QoS at the higher layers.
Conference Paper
Optical Burst Switching (OBS) [1][2][3][4] has been proposed for the next generation Internet based on the Wavelength Division Multiplexing (WDM) [5] network technologies. In the OBS, the serious problem is the burst loss due to the burst contention. There have been a number of traffic control methods [6][7] for solving the burst contention problem. The Internet has two traffic classes, i.e. real-time bursts and reliable bursts. However, most of the conventional traffic controls aren’t suitable for the real-time bursts. In this paper, we propose a new effective traffic control method named a Leaky Bucket Burst Window (LB-BW). The proposed method is composed of the three algorithms, a Leaky Bucket algorithm, a Burst Window algorithm and a determining network state algorithm. Only in a Leaky Bucket algorithm, both real-time bursts and reliable bursts would be delayed under high load. It is possible to reduce the burst loss probability, but real-time bursts are weak in the delay. Therefore by integrating a Burst Window algorithm and a determining network state algorithm, the edge node can transmit real-time burst priority under high load. We show the effectiveness of the proposed method through computer simulations.
Conference Paper
In this paper, we propose an efficient deflection routing method to improve the burst loss performance in Optical Burst Switching (OBS) networks. We show that the proposed scheme gives significant reduction in burst loss rate as compared to conventional technique for optical networks. To improve the burst loss performance, the proposed variable size burst routing algorithm (VSBR), dynamically determines the congestion level and adjusts the burst sending rate, according to the traffic condition in a network.
Article
The FDL buffers can have only discrete delay values. Because of this discontinuity, in order to construct the FDL buffers, some parameters such as the offered load, the average data burst length, and the basic delay unit, of which the length of each FDL is consecutive multiples, should be considered. This means that if one or more parameters change, new FDL buffers are required. So, even when one or more parameters change, in order to minimize the effect of the change, a new service differentiation algorithm dynamically controlling data burst length based on a shared-type feed-forward FDL architecture is proposed in this paper. Various results show that the algorithm improves fairness between classes and significantly reduces the fluctuation of the number of delay lines for each class.
Article
Full-text available
Optical burst switching (OBS) is a promising solution for building terabit optical routers and realizing IP over WDM. In this paper, we describe the basic concept of OBS and present a general architecture of optical core routers and electronic edge routers in the OBS network. The key design issues related to the OBS are also discussed, namely, burst assembly (burstification), channel scheduling, burst offset-time management, and some dimensioning rules. A nonperiodic time-interval burst assembly mechanism is described. A class of data channel scheduling algorithms with void filling is proposed for optical routers using a fiber delay line buffer. The LAUC-VF (latest available unused channel with void filling) channel scheduling algorithm is studied in detail. Initial results on the burst traffic characteristics and on the performance of optical routers in the OBS network with self-similar traffic as inputs are reported in the paper.
Article
Full-text available
In this letter we consider burst switching for very high speed routing in the next generation Internet backbone. In this scenario, Internet Protocol (IP) packets to a given destination are collected in bursts at the network edges. We propose a burst assembly mechanism that can reduce the traffic autocorrelation or degree of self-similarity, and at the same time keep the delay due to burst formation limited at the network edges.
Article
In this paper, we give an overview and classification of optical burst switching schemes and present burst reservation concepts. The performance of various basic reservation mechanisms proposed in literature is compared. Furthermore, a new analysis is introduced that allows to calculate the loss probabilities of a two-class system based on the reservation mechanism just-enough-time (JET) for arbitrary offsets. Finally, a variety of new results is presented including the dependence of burst loss probabilities on offset, burst length distribution, and interarrival distribution.
Article
In this paper, we give an overview and classification of optical burst switching schemes and present burst reservation concepts. The performance of various basic reservation mechanisms proposed in literature is compared. Furthermore, a new analysis is introduced that allows to calculate the loss probabilities of a two-class system based on the reservation mechanism just-enough-time (JET) for arbitrary offsets. Finally, a variety of new results is presented including the dependence of burst loss probabilities on offset, burst length distribution, and interarrival distribution.
Conference Paper
We describe the architecture, performance analysis and simulation result of a novel switching paradigm for optical WDM networks called Just-In-Time Optical Burst Switching (JIT-OBS) designed for ultra-low-latency transport of data-bursts across an optical WDM network. It combines the desirable features of circuit-switching and packet-switching, and features an out-of-band signaling scheme on a separate control channel with explicit feedback on delivery of data-bursts. We provide a performance analysis and simulation of the JITOBS approach, and compare its performance with those of circuit-switching and packet-switching approaches. We find that it has the best latency performance among the different switching mechanisms, and it has a better throughput performance than circuit-switching, and its performance is insensitive to network propagation delays.
Article
We address the issue of how to provide basic quality of service (QoS) in optical burst-switched WDM networks with limited fiber delay lines (FDLs). Unlike existing buffer-based QoS schemes, the novel offset-time-based QoS scheme we study in this paper does not mandate any buffer for traffic isolation, but nevertheless can take advantage of FDLs to improve the QoS. This makes the proposed QoS scheme suitable for the next generation optical Internet. The offset times required for class isolation when making wavelength and FDL reservations are quantified, and the upper and lower bounds on the burst loss probability are analyzed. Simulations are also conducted to evaluate the QoS performance in terms of burst loss probability and queuing delay. We show that with limited FDLs, the offset-time-based QoS scheme can be very efficient in supporting basic QoS.
Article
The rapid pace of development in both Internet applications and emerging optical technologies is bringing about fundamental changes in networking philosophies. Key trends are the emergence of dynamic wavelength provisioning and a corresponding reduction in wavelength provisioning timescales. As this transition continues, the current use of the wavelength-routing paradigm for carrying bursty Internet traffic will likely suffer from various shortcomings associated with circuit-switched networks. Meanwhile, optical packet switching technology is still facing significant cost and technological hurdles. Optical burst switching, or OBS, which represents a balance between circuit and packet switching, has opened up some exciting new dimensions in optical networking. This article describes the OBS paradigm, and also proposes the use of labeled OBS, or LOBS, as a natural control and provisioning solution under the ubiquitous IP multiprotocol label switching framework
Article
To support bursty traffic on the Internet (and especially WWW) efficiently, optical burst switching (OBS) is proposed as a way to streamline both protocol and hardware in building the future generation Optical Internet. By leveraging the attractive properties of optical communications and at the same time, taking into account its limitations, OBS combines the best of optical circuitswitching and packet/cell switching. In this paper, the general concept of OBS protocols and in particular, those based on Just-Enough-Time (JET), is described, along with the applicability of OBS protocols to IP over WDM. Specific issues such as the use of fiber delay-lines (FDL) for accommodating processing delay and/or resolving conflicts are also discussed. In addition, the performance of JET-based OBS protocols which use an offset time along with delayed reservation to achieve efficient utilization of both bandwidth and FDLs as well as to support priority-based routing is evaluated.
QoS performance in IP over WDM networks
  • M Yoo
  • C Qiao