13-CongestioninDataNetworks

49
8/6/2019 13-CongestioninDataNetworks http://slidepdf.com/reader/full/13-congestionindatanetworks 1/49 Data and Computer Data and Computer Communications Communications Eighth Edition Eighth Edition by William Stallings by William Stallings Lecture slides by Lawrie Brown Lecture slides by Lawrie Brown Chapter 13 Chapter 13 ±  ± Congestion in Data Congestion in Data Networks Networks

Transcript of 13-CongestioninDataNetworks

Page 1: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 1/49

Data and Computer Data and Computer 

CommunicationsCommunications

Eighth EditionEighth Editionby William Stallingsby William Stallings

Lecture slides by Lawrie BrownLecture slides by Lawrie Brown

Chapter 13Chapter 13 ± ± Congestion in DataCongestion in DataNetworksNetworks

Page 2: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 2/49

Congestion in Data NetworksCongestion in Data Networks

 At St. Paul's a great throng crammed the platform. She At St. Paul's a great throng crammed the platform. She

 saw a sea of faces, each stamped with a kind of  saw a sea of faces, each stamped with a kind of 

 purposeful, hungry urgency, a determination to get into purposeful, hungry urgency, a determination to get into

this train. As before, when she was on the Northernthis train. As before, when she was on the Northern

 Line, she thought there must be some rule, some Line, she thought there must be some rule, some

operating law, that would stop more than a limited,operating law, that would stop more than a limited,

controlled number getting in. Authority would appear controlled number getting in. Authority would appear and stop it.and stop it.

 ²King Solomon's Carpet , Barbara Vine (Ruth Rendell)

Page 3: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 3/49

What Is Congestion?What Is Congestion?

congestion occurs when the no of packets beingcongestion occurs when the no of packets being

transmitted through the network approaches thetransmitted through the network approaches the

packet handling capacity of the networkpacket handling capacity of the network

congestion control aims to keep no of packetscongestion control aims to keep no of packets

below a level at which performance falls off below a level at which performance falls off 

dramaticallydramatically

a data network is a network of queuesa data network is a network of queues generally 80% utilization is criticalgenerally 80% utilization is critical

finite queues mean data may be lostfinite queues mean data may be lost

Page 4: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 4/49

Queues at a NodeQueues at a Node

Page 5: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 5/49

Interaction of QueuesInteraction of Queues

Page 6: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 6/49

IdealIdeal

NetworkNetworkUtilizationUtilization

Page 7: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 7/49

Effects of Effects of 

CongestionCongestion--

No ControlNo Control

Page 8: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 8/49

Mechanisms for Mechanisms for 

Congestion ControlCongestion Control

Page 9: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 9/49

BackpressureBackpressure

if node becomes congested it can slow down or if node becomes congested it can slow down or halt flow of packets from other nodeshalt flow of packets from other nodes cf. backpressure in blocked fluid pipecf. backpressure in blocked fluid pipe

may mean that other nodes have to apply control onmay mean that other nodes have to apply control on

incoming packet ratesincoming packet rates propagates back to sourcepropagates back to source

can restrict to high traffic logical connectionscan restrict to high traffic logical connections

used in connection oriented nets that allow hopused in connection oriented nets that allow hop

by hop congestion control (eg. X.25)by hop congestion control (eg. X.25) not used in ATM nor frame relaynot used in ATM nor frame relay

only recently developed for IPonly recently developed for IP

Page 10: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 10/49

Choke PacketChoke Packet

a control packeta control packet

generated at congested nodegenerated at congested node

sent to source nodesent to source node

eg. ICMP source quencheg. ICMP source quench

�� from router or destinationfrom router or destination

�� source cuts back until no more source quench messagesource cuts back until no more source quench message

�� sent for every discarded packet, or anticipatedsent for every discarded packet, or anticipated

is a rather crude mechanismis a rather crude mechanism

Page 11: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 11/49

Implicit Congestion SignalingImplicit Congestion Signaling

transmission delay increases with congestiontransmission delay increases with congestion

hence a packet may be discardedhence a packet may be discarded

source detects this implicit congestion indicationsource detects this implicit congestion indication useful on connectionless (datagram) networksuseful on connectionless (datagram) networks

eg. IP basedeg. IP based

�� (TCP includes congestion and flow control(TCP includes congestion and flow control -- see chapter 17)see chapter 17)

used in frame relay LAPFused in frame relay LAPF

Page 12: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 12/49

Explicit Congestion SignalingExplicit Congestion Signaling

network alerts end systems of increasingnetwork alerts end systems of increasing

congestioncongestion

end systems take steps to reduce offered loadend systems take steps to reduce offered load

BackwardsBackwards

congestion avoidance notification in opposite directioncongestion avoidance notification in opposite direction

to packet requiredto packet required

ForwardsForwards congestion avoidance notification in same direction ascongestion avoidance notification in same direction as

packet requiredpacket required

Page 13: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 13/49

Explicit Signaling CategoriesExplicit Signaling Categories

BinaryBinary

a bit set in a packet indicates congestiona bit set in a packet indicates congestion

Credit basedCredit based indicates how many packets source may sendindicates how many packets source may send

common for end to end flow controlcommon for end to end flow control

Rate basedRate based supply explicit data rate limitsupply explicit data rate limit

nodes along path may request rate reductionnodes along path may request rate reduction

eg. ATMeg. ATM

Page 14: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 14/49

Traffic ManagementTraffic Management

fairnessfairness

provide equal treatment of various flowsprovide equal treatment of various flows

quality of servicequality of service different treatment for different connectionsdifferent treatment for different connections

reservationsreservations

traffic contract between user and networktraffic contract between user and network carry bestcarry best--effort or discard excess trafficeffort or discard excess traffic

Page 15: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 15/49

Congestion Control in PacketCongestion Control in Packet

Switched NetworksSwitched Networks send control packet to some or all sourcesend control packet to some or all source

nodesnodes

requires additional traffic during congestionrequires additional traffic during congestion

rely on routing informationrely on routing information

may react too quicklymay react too quickly

end to end probe packetsend to end probe packets

adds to overheadadds to overhead

add congestion info to packets in transitadd congestion info to packets in transit

either backwards or forwardseither backwards or forwards

Page 16: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 16/49

Frame RelayFrame Relay

Congestion ControlCongestion Control minimize discardsminimize discards

maintain agreed QoSmaintain agreed QoS

minimize probability of one end user monopolyminimize probability of one end user monopoly

simple to implementsimple to implement

create minimal additional trafficcreate minimal additional traffic

distribute resources fairlydistribute resources fairly

limit spread of congestionlimit spread of congestion operate effectively regardless of traffic flowoperate effectively regardless of traffic flow

minimum impact on other systemsminimum impact on other systems

minimize variance in QoSminimize variance in QoS

Page 17: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 17/49

FR Control TechniquesFR Control Techniques

difficult for framedifficult for frame--relayrelay

 joint network & end joint network & end--system responsibilitysystem responsibility

techniques:techniques: discard strategydiscard strategy

congestion avoidancecongestion avoidance

explicit signalingexplicit signaling

congestion recoverycongestion recovery

implicit signaling mechanismimplicit signaling mechanism

Page 18: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 18/49

Traffic Rate ManagementTraffic Rate Management

must discard frames to cope with congestionmust discard frames to cope with congestion

arbitrarily, no regard for sourcearbitrarily, no regard for source

no reward for restraint so end systems transmit asno reward for restraint so end systems transmit as

fast as possiblefast as possible Committed information rate (CIR)Committed information rate (CIR)

�� data in excess of this liable to discarddata in excess of this liable to discard

�� not guaranteed in extreme congestion situationsnot guaranteed in extreme congestion situations

�� aggregate CIR should not exceed physical data rateaggregate CIR should not exceed physical data rate

Committed burst sizeCommitted burst size

Excess burst sizeExcess burst size

Page 19: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 19/49

Operation of CIROperation of CIR

Page 20: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 20/49

RelationshipRelationship

AmongAmongCongestionCongestion

ParametersParameters

Page 21: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 21/49

Congestion Avoidance usingCongestion Avoidance using

Explicit Signaling

Explicit Signaling

network alerts end systems of growingnetwork alerts end systems of growing

congestion usingcongestion using

backward explicit congestion notificationbackward explicit congestion notification forward explicit congestion notificationforward explicit congestion notification

frame handler monitors its queuesframe handler monitors its queues

may notify some or all logical connectionsmay notify some or all logical connections user response reduce rateuser response reduce rate

Page 22: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 22/49

ATM Traffic ManagementATM Traffic Management

high speed, small cell size, limited overhead bitshigh speed, small cell size, limited overhead bits

still evolvingstill evolving

reasons existing tools are inadequate for ATMreasons existing tools are inadequate for ATM majority of traffic not amenable to flow controlmajority of traffic not amenable to flow control

feedback slow due to reduced transmission timefeedback slow due to reduced transmission timecompared with propagation delaycompared with propagation delay

wide range of application demandswide range of application demands

different traffic patternsdifferent traffic patterns different network servicesdifferent network services

high speed switching and transmission increaseshigh speed switching and transmission increasesvolatilityvolatility

Page 23: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 23/49

Latency/Speed EffectsLatency/Speed Effects

consider ATM at 150Mbpsconsider ATM at 150Mbps

takes ~2.8x10takes ~2.8x10--66 seconds to insert single cellseconds to insert single cell

time to traverse network depends ontime to traverse network depends on

propagation delay and switching delaypropagation delay and switching delay

assume propagation at twoassume propagation at two--thirds speed of lightthirds speed of light

if source and destination on opposite sides of if source and destination on opposite sides of USA, propagation time ~ 48x10USA, propagation time ~ 48x10--33 secondsseconds

given implicit congestion control, by the timegiven implicit congestion control, by the timedropped cell notification has reached source,dropped cell notification has reached source,7.2x107.2x1066 bits have been transmittedbits have been transmitted

this is not a good strategy for ATMthis is not a good strategy for ATM

Page 24: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 24/49

Cell Delay VariationCell Delay Variation

for ATM voice/video, data is a stream of cellsfor ATM voice/video, data is a stream of cells

delay across network must be shortdelay across network must be short

rate of delivery must be constantrate of delivery must be constant there will always be some variation in transitthere will always be some variation in transit

delay cell delivery to application so that constantdelay cell delivery to application so that constant

bit rate can be maintained to applicationbit rate can be maintained to application

Page 25: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 25/49

Timing of CBR CellsTiming of CBR Cells

Page 26: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 26/49

Network Contribution toNetwork Contribution to

Cell Delay VariationCell Delay Variation in packet switched networks is due toin packet switched networks is due to

queuing delays and routing decision timequeuing delays and routing decision time

in Frame relay networks is similar in Frame relay networks is similar 

in ATM networksin ATM networks

less than frame relayless than frame relay

 ATM protocol designed to minimize ATM protocol designed to minimize

processing overheads at switchesprocessing overheads at switches  ATM switches have very high throughput ATM switches have very high throughput

only noticeable delay is from congestiononly noticeable delay is from congestion

must not accept load that causes congestionmust not accept load that causes congestion

Page 27: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 27/49

Cell Delay VariationCell Delay Variation

AtT

he UNIAtT

he UNI application produces data at fixed rateapplication produces data at fixed rate

3 layers of ATM processing causes delay3 layers of ATM processing causes delay

interleaving cells from different connectionsinterleaving cells from different connections

operation and maintenance cell interleavingoperation and maintenance cell interleaving

if using synchronous digital hierarchy frames,if using synchronous digital hierarchy frames,

these are inserted at physical layer these are inserted at physical layer  cannot predict these delayscannot predict these delays

Page 28: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 28/49

Origins of Cell Delay VariationOrigins of Cell Delay Variation

Page 29: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 29/49

Traffic and CongestionTraffic and Congestion

ControlF

rameworkControlF

ramework ATM layer traffic and congestion control ATM layer traffic and congestion control

should support QoS classes for allshould support QoS classes for allforeseeable network servicesforeseeable network services

should not rely on AAL protocols that areshould not rely on AAL protocols that arenetwork specific, nor higher levelnetwork specific, nor higher levelapplication specific protocolsapplication specific protocols

should minimize network and end to endshould minimize network and end to endsystem complexitysystem complexity

Page 30: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 30/49

Timings ConsideredTimings Considered

timing intervals considered:timing intervals considered: cell insertion timecell insertion time

round trip propagation timeround trip propagation time

connection durationconnection duration long termlong term

traffic control strategy then must:traffic control strategy then must: determine whether a given new connection can bedetermine whether a given new connection can be

accommodatedaccommodated agree performance parameters with subscriber agree performance parameters with subscriber 

now review various control techniquesnow review various control techniques

Page 31: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 31/49

Resource Management UsingResource Management Using

Virtual PathsVirtual Paths separate traffic flow according to serviceseparate traffic flow according to service

characteristics on a virtual pathcharacteristics on a virtual path

user to user applicationuser to user application

user to network applicationuser to network application

network to network applicationnetwork to network application

QoS parameters concerned with are:QoS parameters concerned with are:

cell loss ratiocell loss ratio cell transfer delaycell transfer delay

cell delay variationcell delay variation

Page 32: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 32/49

Configuration of Configuration of 

VCCs and VPCsVCCs and VPCs

Page 33: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 33/49

Allocating VCCs within VPCAllocating VCCs within VPC

all VCCs within VPC should experienceall VCCs within VPC should experiencesimilar network performancesimilar network performance

options for allocation:options for allocation:

aggregate peak demandaggregate peak demand�� set VPC capacity to total of all peak VCC ratesset VPC capacity to total of all peak VCC rates

�� will meet peak demands, but often underutilizedwill meet peak demands, but often underutilized

statistical multiplexingstatistical multiplexing�� set VPC capacity to more than average VCC ratesset VPC capacity to more than average VCC rates

�� will see greater variation but better utilizationwill see greater variation but better utilization

Page 34: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 34/49

Connection AdmissionConnection Admission

ControlControl first line of defensefirst line of defense

user specifies traffic characteristics for newuser specifies traffic characteristics for newconnection (VCC or VPC) by selecting a QoSconnection (VCC or VPC) by selecting a QoS

network accepts connection only if it can meetnetwork accepts connection only if it can meetthe demandthe demand

traffic contracttraffic contract peak cell ratepeak cell rate

cell delay variationcell delay variation sustainable cell ratesustainable cell rate

burst toleranceburst tolerance

Page 35: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 35/49

Usage Parameter ControlUsage Parameter Control

UPC function monitors a connection toUPC function monitors a connection to

ensure traffic obeys contractensure traffic obeys contract

purpose is to protect network resourcespurpose is to protect network resourcesfrom overload by one connectionfrom overload by one connection

done on VCC and VPCdone on VCC and VPC

peak cell rate and cell delay variationpeak cell rate and cell delay variation sustainable cell rate and burst tolerancesustainable cell rate and burst tolerance

UPC discards cells outside traffic contractUPC discards cells outside traffic contract

Page 36: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 36/49

Selective Call DiscardSelective Call Discard

when network at point beyond UPCwhen network at point beyond UPC

discards (CLP=1) cellsdiscards (CLP=1) cells

aim to discard lower aim to discard lower--priority cells whenpriority cells whencongested to protect higher congested to protect higher--priority cellspriority cells

note. can¶t distinguish between cells originallynote. can¶t distinguish between cells originally

labeled lower priority, verses those tagged bylabeled lower priority, verses those tagged by

UPC functionUPC function

Page 37: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 37/49

Traffic ShapingTraffic Shaping

UPC provides a form of traffic policingUPC provides a form of traffic policing

can be desirable to also shape trafficcan be desirable to also shape traffic

smoothing out traffic flowsmoothing out traffic flow reducing cell clumpingreducing cell clumping

token buckettoken bucket

Page 38: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 38/49

Token BucketToken Bucket for for 

Traffic ShapingT

raffic Shaping

Page 39: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 39/49

GFR Traffic ManagementGFR Traffic Management

guaranteed frame rate (GFR) as simple as UBRguaranteed frame rate (GFR) as simple as UBRfrom end system viewpointfrom end system viewpoint

places modest requirements on ATM networkplaces modest requirements on ATM network

end system does no policing or shaping of trafficend system does no policing or shaping of traffic may transmit at line rate of ATM adaptor may transmit at line rate of ATM adaptor 

no guarantee of deliveryno guarantee of delivery so higher layer (eg. TCP) must do congestion controlso higher layer (eg. TCP) must do congestion control

user can reserve capacity for each VCuser can reserve capacity for each VC ensures application can send at min rate with no lossensures application can send at min rate with no loss

if no congestion, higher rates maybe usedif no congestion, higher rates maybe used

Page 40: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 40/49

Frame RecognitionFrame Recognition

GFR recognizes frames as well as cellsGFR recognizes frames as well as cells

when congested, network discards whole framewhen congested, network discards whole frame

rather than individual cellsrather than individual cells

all cells of a frame have same CLP bit settingall cells of a frame have same CLP bit setting

CLP=1 AAL5 frames lower priority (best effort)CLP=1 AAL5 frames lower priority (best effort)

CLP=0 frames minimum guaranteed capacityCLP=0 frames minimum guaranteed capacity

Page 41: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 41/49

GFR Contract ParametersGFR Contract Parameters

Peak cell rate (PCR)Peak cell rate (PCR)

Minimum cell rate (MCR)Minimum cell rate (MCR)

Maximum burst size (MBS)Maximum burst size (MBS)Maximum frame size (MFS)Maximum frame size (MFS)

Cell delay variation tolerance (CDVT)Cell delay variation tolerance (CDVT)

Page 42: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 42/49

Components of GFR SystemComponents of GFR System

Supporting RateG

uaranteesSupporting RateG

uarantees

Page 43: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 43/49

Tagging and PolicingTagging and Policing

discriminates between frames thatdiscriminates between frames that

conform to contract and those that don¶tconform to contract and those that don¶t

set CLP=1 on all cells in frame if notset CLP=1 on all cells in frame if not gives lower prioritygives lower priority

maybe done by network or sourcemaybe done by network or source

network may discard CLP=1 cellsnetwork may discard CLP=1 cells policingpolicing

Page 44: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 44/49

Buffer ManagementBuffer Management

deals with treatment of buffered cellsdeals with treatment of buffered cells

congestion indicated by high buffer congestion indicated by high buffer occupancyoccupancy

will discard tagged cells in preference towill discard tagged cells in preference tountagged cellsuntagged cells

including ones already in buffer to make roomincluding ones already in buffer to make room

may do per VC buffering for fairnessmay do per VC buffering for fairness cell discard based on queuecell discard based on queue--specificspecific

thresholdsthresholds

Page 45: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 45/49

SchedulingScheduling

preferential treatment to untagged cellspreferential treatment to untagged cells

separate queues for each VCseparate queues for each VC

make per make per--VC scheduling decisionsVC scheduling decisions enables control of outgoing rate of VCsenables control of outgoing rate of VCs

VCs get fair capacity allocationVCs get fair capacity allocation

still meet contractstill meet contract

Page 46: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 46/49

GFC Conformance DefinitionGFC Conformance Definition

UPC function monitors each active VCUPC function monitors each active VC

to ensure traffic conforms to contractto ensure traffic conforms to contract

tag or discard nonconforming cellstag or discard nonconforming cells

frame conforms if all cells conformframe conforms if all cells conform

a cell conforms if:a cell conforms if:

rate of cells is within contractrate of cells is within contract

all cells in frame have same CLPall cells in frame have same CLP frame satisfies MFS parameter frame satisfies MFS parameter 

�� check if either last cell in frame or cell count < MFScheck if either last cell in frame or cell count < MFS

Page 47: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 47/49

QoS Eligibility TestQoS Eligibility Test

two stage filtering processtwo stage filtering process

a frame is tested for conformance to contracta frame is tested for conformance to contract if not, may discard or tagif not, may discard or tag

set upper bound & penalize cells above upper boundset upper bound & penalize cells above upper bound do expect attempt to deliver tagged cellsdo expect attempt to deliver tagged cells

determine frames eligible for QoS guaranteesdetermine frames eligible for QoS guarantees under GFR contract for VCunder GFR contract for VC

set lower bound on trafficset lower bound on traffic frames in traffic flow below threshold are eligibleframes in traffic flow below threshold are eligible

Page 48: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 48/49

GFR VC Frame CategoriesGFR VC Frame Categories

nonconforming framenonconforming frame

cells of this frame will be tagged or discardedcells of this frame will be tagged or discarded

conforming but ineligible framesconforming but ineligible frames cells will receive a bestcells will receive a best--effort serviceeffort service

conforming and eligible framesconforming and eligible frames

cells will receive a guarantee of deliverycells will receive a guarantee of delivery

form of cell rate algorithm is usedform of cell rate algorithm is used

Page 49: 13-CongestioninDataNetworks

8/6/2019 13-CongestioninDataNetworks

http://slidepdf.com/reader/full/13-congestionindatanetworks 49/49

SummarySummary

congestion effectscongestion effects

congestion controlcongestion control

traffic managementtraffic management frame relay congestion controlframe relay congestion control

 ATM congestion control ATM congestion control