15-441 Computer NetworkingOverviewQueuing DisciplinesTypical Internet QueuingFIFO + Drop-tail ProblemsSlide 6Active Queue ManagementDesign ObjectivesLock-out ProblemFull Queues ProblemRandom Early Detection (RED)RED AlgorithmRED OperationSlide 14Fairness GoalsWhat is Fairness?Max-min FairnessImplementing Max-min FairnessBit-by-bit RRBit-by-bit RR IllustrationFair QueuingFQ IllustrationBit-by-bit RR ExampleFair Queuing Tradeoffs15-441 Computer NetworkingLecture 17 – Queue ManagementAs usual: Thanks to Srini Seshan and Dave AndersonLecture 22: 2006-11-14 2Overview•Queue management & RED•Fair-queuingLecture 22: 2006-11-14 3Queuing Disciplines•Each router must implement some queuing discipline•Queuing allocates both bandwidth and buffer space:•Bandwidth: which packet to serve (transmit) next •Buffer space: which packet to drop next (when required)•Queuing also affects latencyLecture 22: 2006-11-14 4Typical Internet Queuing•FIFO + drop-tail•Simplest choice•Used widely in the Internet•FIFO (first-in-first-out) •Implies single class of traffic•Drop-tail•Arriving packets get dropped when queue is full regardless of flow or importance•Important distinction:•FIFO: scheduling discipline•Drop-tail: drop policyLecture 22: 2006-11-14 5FIFO + Drop-tail Problems•Leaves responsibility of congestion control completely to the edges (e.g., TCP)•Does not separate between different flows•No policing: send more packets get more service•Synchronization: end hosts react to same eventsLecture 22: 2006-11-14 6FIFO + Drop-tail Problems•Full queues•Routers are forced to have have large queues to maintain high utilizations•TCP detects congestion from loss•Forces network to have long standing queues in steady-state•Lock-out problem•Drop-tail routers treat bursty traffic poorly•Traffic gets synchronized easily allows a few flows to monopolize the queue spaceLecture 22: 2006-11-14 7Active Queue Management•Design active router queue management to aid congestion control •Why?•Router has unified view of queuing behavior•Routers see actual queue occupancy (distinguish queue delay and propagation delay)•Routers can decide on transient congestion, based on workloadLecture 22: 2006-11-14 8Design Objectives•Keep throughput high and delay low•High power (throughput/delay)•Accommodate bursts•Queue size should reflect ability to accept bursts rather than steady-state queuing•Improve TCP performance with minimal hardware changesLecture 22: 2006-11-14 9Lock-out Problem•Random drop•Packet arriving when queue is full causes some random packet to be dropped•Drop front•On full queue, drop packet at head of queue•Random drop and drop front solve the lock-out problem but not the full-queues problemLecture 22: 2006-11-14 10Full Queues Problem•Drop packets before queue becomes full (early drop)•Intuition: notify senders of incipient congestion•Example: early random drop (ERD):•If qlen > drop level, drop each new packet with fixed probability p•Does not control misbehaving usersLecture 22: 2006-11-14 11Random Early Detection (RED)•Detect incipient congestion•Assume hosts respond to lost packets•Avoid window synchronization•Randomly mark packets•Avoid bias against bursty trafficLecture 22: 2006-11-14 12RED Algorithm•Maintain running average of queue length•If avg < minth do nothing•Low queuing, send packets through•If avg > maxth, drop packet•Protection from misbehaving sources•Else mark packet in a manner proportional to queue length•Notify sources of incipient congestionLecture 22: 2006-11-14 13RED OperationMin threshMax threshAverage Queue LengthminthmaxthmaxP1.0Avg queue lengthP(drop)Lecture 22: 2006-11-14 14Overview•Queue management & RED•Fair-queuingLecture 22: 2006-11-14 15Fairness Goals•Allocate resources fairly •Isolate ill-behaved users•Router does not send explicit feedback to source•Still needs e2e congestion control•Still achieve statistical muxing•One flow can fill entire pipe if no contenders•Work conserving scheduler never idles link if it has a packetLecture 22: 2006-11-14 16What is Fairness?•At what granularity?•Flows, connections, domains?•What if users have different RTTs/links/etc.•Should it share a link fairly or be TCP fair?•Maximize fairness index?•Fairness = ( xi)2/n(xi2) 0<fairness<1•Basically a tough question to answer – typically design mechanisms instead of policy•User = arbitrary granularityLecture 22: 2006-11-14 17Max-min Fairness•Allocate user with “small” demand what it wants, evenly divide unused resources to “big” users•Formally:•Resources allocated in terms of increasing demand•No source gets resource share larger than its demand•Sources with unsatisfied demands get equal share of resourceLecture 22: 2006-11-14 18Implementing Max-min Fairness•Generalized processor sharing•Fluid fairness•Bitwise round robin among all queues•Why not simple round robin?•Variable packet length can get more service by sending bigger packets•Unfair instantaneous service rate•What if arrive just before/after packet departs?Lecture 22: 2006-11-14 19Bit-by-bit RR•Single flow: clock ticks when a bit is transmitted. For packet i:•Pi = length, Ai = arrival time, Si = begin transmit time, Fi = finish transmit time•Fi = Si+Pi = max (Fi-1, Ai) + Pi•Multiple flows: clock ticks when a bit from all active flows is transmitted round number•Can calculate Fi for each packet if number of flows is know at all times•Why do we need to know flow count? need to know A This can be complicatedLecture 22: 2006-11-14 20Bit-by-bit RR Illustration•Not feasible to interleave bits on real networks•FQ simulates bit-by-bit RRLecture 22: 2006-11-14 21Fair Queuing•Mapping bit-by-bit schedule onto packet transmission schedule•Transmit packet with the lowest Fi at any given time•How do you compute Fi?Lecture 22: 2006-11-14 22FQ IllustrationFlow 1Flow 2Flow nI/PO/PVariation: Weighted Fair Queuing (WFQ)Lecture 22: 2006-11-14 23Bit-by-bit RR ExampleF=10Flow 1(arriving)Flow 2transmittingF=2OutputF=5F=8Flow 1 Flow 2OutputF=10Cannot preempt packetcurrently being transmittedLecture 22: 2006-11-14 24Fair Queuing Tradeoffs•FQ can control congestion by monitoring flows•Non-adaptive flows can still be a problem – why?•Complex state•Must keep queue per flow•Hard in routers with many flows (e.g., backbone
View Full Document