Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more ➡
Download
Standard view
Full view
of .
Add note
Save to My Library
Sync to mobile
Look up keyword
Like this
1Activity
×
0 of .
Results for:
No results containing your search query
P. 1
Comparison between Different Scheduling Strategies by Using Cost239 Optical Network

Comparison between Different Scheduling Strategies by Using Cost239 Optical Network

Ratings: (0)|Views: 103|Likes:
In this paper we present different demand policies in scheduled lightpath demand (SLDs) .SLD is a demand for a set of lightpaths (connections), defined by a tuple (s,d,n,α,ω) where s and d are the source and destination nodes of the lightpaths, n is the number of requested lightpaths α, ω and are the set-up and tear-down times of the lightpaths. The objective of this paper to increase resource utilization ratio by using channel reuse. In this paper we works on Cost 239 network by assigning the same channel to several lightpaths, by using different wavelengths which varies according to the time (set-up and tear-down).By comparison of demand policies which also depend upon time(set-up and tear-down) every demand policy has own schedule. By using Cost 239 network we found the result which demand policy is effective and scheduled first and we compare each policy by using graphical representation and then find which policy is best for scheduling and increase the resource utilization .
In this paper we present different demand policies in scheduled lightpath demand (SLDs) .SLD is a demand for a set of lightpaths (connections), defined by a tuple (s,d,n,α,ω) where s and d are the source and destination nodes of the lightpaths, n is the number of requested lightpaths α, ω and are the set-up and tear-down times of the lightpaths. The objective of this paper to increase resource utilization ratio by using channel reuse. In this paper we works on Cost 239 network by assigning the same channel to several lightpaths, by using different wavelengths which varies according to the time (set-up and tear-down).By comparison of demand policies which also depend upon time(set-up and tear-down) every demand policy has own schedule. By using Cost 239 network we found the result which demand policy is effective and scheduled first and we compare each policy by using graphical representation and then find which policy is best for scheduling and increase the resource utilization .

More info:

categoriesTypes, Research
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See More
See less

03/03/2014

pdf

text

original

 
 International Journal of Advanced Computer Research (ISSN (print): 2249-7277 ISSN (online): 2277-7970)Volume-2 Number-4 Issue-7 December-2012
 
125
Comparison between Different Scheduling Strategies by Using Cost239Optical Network
Pooja Meena
1
, Manish Shrivastava
2
, Sushil Chaturvedi
3
Mtech Scholar,
 
Dept. Information Technology,
LNCT, BHOPAL, M.P., INDIA
1
Head of the department, Dept. Information Technology,
LNCT, BHOPAL, M.P., INDIA
2
Associate professor, Dept. Information Technology,
LNCT, BHOPAL, M.P., INDIA
3
 
1
meena.pooja1@gmail.com
2
maneesh.shreevastava@yahoo.com
3
sushil524@gmail.com
Abstract
 In this paper we present different demand policiesin scheduled lightpath demand (SLDs) .SLD is a demand for a set of lightpaths (connections), defined by a tuple
(s,d,n,α,ω
 ) where s and d are the source and destination nodes of the lightpaths, n is the number of requested lightpaths
α, ω
and are the set-up and tear-down times of the lightpaths. The objective of this paper to increase resourceutilization ratio by using channel reuse. In this paper we works on Cost 239 network by assigning the same channel to several lightpaths, by using different wavelengths which varies according to the time (set-up and tear-down).By comparison of  demand policies which also depend upon time(set-up and tear-down) every demand policy has own schedule. By using Cost 239 network we found the result which demand policy is effective and  scheduled first and we compare each policy byusing graphical representation
 
 and then find which policy is best for scheduling and increase the resource utilization .
Keywords
WDM, RWA, STATIC TRAFFIC, DEMAND POLICIES
 
1. Introduction
At the present scenario the popularity of the Internetand using of Internet exponential increasing day byday
in computer society that’s why
bandwidthdemands also increasing. Wavelength divisionmultiplexing (WDM) divides the enormousbandwidth of an optical fiber into many non-overlapping wavelength channels. WDM opticalnetworks are able to meet the rapid growth of bandwidth demands and are considered to be themost appropriate choice for future Internet backbone.In a WDM optical network with path basedprotection, each connection has two link disjointpaths: one working path and one protection path. Aconnection request is considered blocked if nosufficient resources are available to route either path.Efficient resource utilization can be achieved throughbackup resource sharing, a technique that allowsmultiple protection paths to share some commonwavelength-links as long as their correspondingworking paths are link disjoint while 100%restorability is still maintained. Shared pathprotection outperforms, in terms of resource used,other protection techniques based on the dedicatedreservation of backup capacities [1]. In Opticalnetwork optoelectronic approach mainly used,Optoelectronic approach is the most practical methodtoday to realize wavelength conversion [4].
 1.1 RWA Problem
In informal terms, the RWA problem underconsideration is the following: given a network and aset of SLDs, find the routing in the network for theSLDs which meets an optimality criterion. Then, forthis routing solution, find the assignment of wavelengths to lightpaths that minimizes the numberof required wavelengths, while satisfying thewavelength continuity constraint (no wavelengthconversion exists in the optical network). Differentoptimality criteria may be considered in the routingsub-problem, for instance, the minimization of thenumber of WDM channels or the minimization of thecongestion (the number of lightpaths in the mostloaded link).congestion minimization is importantwhen the number of wavelengths in the network islimited [3].In wavelength convertible networks, a lightpath maybe assigned a different channel on each fiber ittraverses. However, full range all-optical wavelengthconversion is generally not feasible, due to both costand technological restrictions. Therefore, most
 
 International Journal of Advanced Computer Research (ISSN (print): 2249-7277 ISSN (online): 2277-7970)Volume-2 Number-4 Issue-7 December-2012
 
126practical networks do not assume wavelengthconversion capabilities. In the absence of wavelengthconverters, a lightpath must be assigned the samechannel on all links. This is called the wavelengthcontinuity constraint. The RWA problem, under boththe static and dynamic traffic models. In recent yearsthere has been an increasing number of applicationsthat require periodic use of lightpaths (e.g. once perday, or once per week) at predefined times. For
example, an online “class”
with one two hour lectureper week on a specified day and at a specified time,or a bank transferring its data to a central locationevery night between 2 am and 4 am. A new model,called the scheduled traffic model, has been proposedin the literature to handle such demands [2].Thetraffic model based on SLDs is different from thestatic and dynamic-random lightpath traffic models.Static traffic means that all the demands are known inadvance and do not change over time, whereasdynamic-random traffic means that the arrival andholding times of demands are random. The SLDtraffic model is deterministic because all the demandsare known in advance and is dynamic
 
because it takesinto account the evolution of the traffic load in thenetwork over time. In this paper we are using statictraffic, because it is fixed and do not change overtime [3]. In this paper, we are introducing differentdemand policies in scheduled traffic model by usingstatic lightpath traffic .Our objective is to maximizeresource utilization by using channel reuse method,here also proposed heuristics algorithm of demandpolicies and by using different network comparewhich policy is best for scheduling the request.
2. Related work
2.1 Scheduled Lightpath Demand
SLD means how many requests arrive on a givennetwork in a given amount of time.
IN SLDs’
according to set-up and tear-down time requestsarrived .SLD is a demand for a set of lightpaths(connections), defined by a tuple
(s,d,n,α,ω
) where sand d are the source and destination nodes of thelightpaths, n is the number of requested lightpaths
α,ω
and are the set-up and tear-down times of thelightpaths. In SLD on a single lightpath more thanone no of requests has arrived, if request rate ishigher than the nominal rate of single lightpath [5].Tominimize the amount of channels required toinstantiate the demanded lightpaths. Indeed, a samechannel can be allocated to multiple lightpaths.Clearly, the more a channel is shared, the smaller thetotal amount of required channels is [5]. In recentyears there has been an increasing number of applications that require periodic use of lightpaths(e.g. once per day, or once per week) at predefinedtimes. For example, an online
“class”
with one twohour lecture per week on a specified day and at aspecified time, or a bank transferring its data to acentral location every night between 2 am and 4am..[2]
2.2 Traffic ModelsStatic traffic:-
 The traffic model based on SLDs is different from thestatic and dynamic-random lightpath traffic models.Static traffic means that all the demands are knownin advance and do not change over time, whereasdynamic-random traffic means that the arrival andholding times of demands are random. The SLDtraffic model is deterministic
 
because all the demandsare known in advance and is dynamic
 
because it takesinto account the evolution of the traffic load in thenetwork over time [3].In this paper we are workingon Static Traffic, it is fixed and do not change overtime .In static traffic all the demands known inadvance .In various networks and their requests(no.of request arrived and no. of request accepted ) if timing (set-up and tear-down) is fixed then we canreduce the blocking probability(It is defined as thenumber of connections rejected to the total number of 
 
connection request arrived ,this case number of wavelength per fiber-link is assumed to be fixed) andresource utilization ratio(It is the ratio of total
 
network capacity used to the number of requests areaccepted in the network ) decreased [3].
3. Demand Scheduling Policies
Demand based on chosen scheduling policies whichresult in a scheduling order of demands. Schedulingalways based on set-up and tear-down time of demand (set-up time when request is arrive and tear-down time, when the request is reached betweensource to destination).Different scheduling policiesare proposed below and their impact will alsostudied.
3.1 Earliest-setup Demand First (ESDF):-
This policy schedules demands in increasingorder of their set-up time.The demand with the same set-up time ,thedemand with an earlier tear-down time willbe scheduled first.[6]In Earliest-setup Demand First all the requests whicharrived, we arrange them increasing order of their set-up time. It means, which request come earlier we putthat request in first, then another bigger than that inthis type process will be scheduled.
 
 International Journal of Advanced Computer Research (ISSN (print): 2249-7277 ISSN (online): 2277-7970)Volume-2 Number-4 Issue-7 December-2012
 
127For example: there are 4 requests (with their set-upand tear-down time) given.First request 4 8Second request 3 5Third request 4 10Fourth request 2 4Then we arrange them requests as according to theirset-up time in increasing order, so arrangement willbe belowSet-up time: - 2 3 4 4
3.2 Earliest-Teardown Demand First (ESDF):-
 This policy schedules demands in increasingorder of their tear-down time.The earlier demand ends, the earlier it willbe scheduled.The demand with the same tear-down time,the demand with an earlier set-up time willbe scheduled first.[6]For example: there are 4 requests (with their set-upand tear-down time) given.First request 4 8Second request 3 5Third request 4 10Fourth request 2 4Then we arrange them requests as according to theirtear-down time in decreasing order, so arrangementwill be belowTear-down time: 10 8 5 4
3.3 Most Conflicting Demand First (MCDF):-
This policy schedules demands indecreasing order of their time conflictindices.That is more demand overlaps in time withother demand the earlier it will be processed.for the same time conflict index policiesESDF and ETDF will be applied in order.[6]In this demand policy ,if 4 requests is giventhen according to their timing(set-up andtear-down)we compare each request andfind which request is conflict more toanother request then ,that request will beprocess first than another conflict requestsprocess as order of their confliction.For example: there are 4 requests (with their set-upand tear-down time) given.First request 4 7Second request 3 5Third request 4 6Fourth request 2 4Here, we compare all requests each other and findthat request 2
nd
(3,5)is conflicted more than another
requests ,that’s why request 2
nd
(3,5) request willschedule first another requests will also schedule.
3.4 Least Conflicting Demand First (LCDF):-
In contrast to MCDF ,this policy schedulesdemand in increasing order of their timeconflict indicesIn LCDF ,which demand conflicting less,first schedule that demand, then afteranother less conflicting demand will bescheduling[6]For example: there are 4 requests (with their set-upand tear-down time) given.First request 3 8Second request 1 6Third request 2 5Fourth request 7 10Here, we compare all requests each other and findthat request 4
th
(7,10)is conflicted less than another
requests ,that’s why
request 4
th
(7,10) request willschedule first another requests will also schedule.This is the process how Demand policies are working
4. Experimental Evaluation
In this section, we first evaluate the performance of the proposed demand policies algorithm described inprevious section by comparing it with COST239network .COST239 network has 11 nodes and 23links and all the nodes connected in the network bybidirectional links, to reach one node to another node,we can choose many routes, but we follow theshortest route path for remove the congestion (It isdefined as the maximum wavelength used on the link in the network. For the good scheduling policy thevalue of all the network parameters should beminimum), that the reason behind We have alwayschoose that route which has less congestion by usingmore available wavelengths
.
In Cost 239 network by

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->