CS640: Introduction to Computer NetworksThe Road AheadQueuing DisciplinesTypical Internet QueuingFIFO + Drop-tail ProblemsSlide 6Active Queue ManagementLock-out ProblemFull Queues ProblemRandom Early Detection (RED)RED AlgorithmRED OperationFair Queuing: GoalsWhat is “Fairness”?Max-min FairnessImplementing Max-min FairnessBit-by-bit RRBit-by-bit RR IllustrationFair QueuingFQ IllustrationBit-by-bit RR ExampleFair Queuing TradeoffsToken Bucket for Traffic PolicingToken Bucket CharacteristicsCS640: Introduction to Computer NetworksAditya AkellaLecture 20 -Queuing and Basics of QoS2The Road Ahead•Queuing Disciplines•Fair Queuing•Token Bucket3Queuing Disciplines•Each router must implement some queuing discipline–Scheduling discipline–Drop policy•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 latency•Important for QoS; also for best effort4Typical Internet Queuing•FIFO + drop-tail–Simplest choice–Used widely in the Internet–FIFO: scheduling discipline–Drop-tail: drop policy•FIFO (first-in-first-out) –Implies single class of traffic, no priority•Drop-tail–Arriving packets get dropped when queue is full regardless of flow or importance5•Lock-out problem–Drop-tail routers treat bursty traffic poorly–Traffic gets synchronized easily allows a few flows to monopolize the queue space•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-stateFIFO + Drop-tail Problems6FIFO + Drop-tail Problems•No policing: send more packets get more service–Lack of isolation among flows•Synchronization: end hosts react to same events–Full queue empty Full empty…•Poor support for bursty traffic–Almost always see burst losses!7Active Queue Management•Design “active” router queue management to facilitate better behavior under congestion•Objectives: solve FIFO problems, better support for QoS–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–Research focus: Improve TCP performance with minimal hardware changes8Lock-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 problem9Full 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 users10Random Early Detection (RED)•Detect incipient congestion•Assume hosts respond to lost packets–Compliant congestion control•Avoid window synchronization–Randomly mark packets•Avoid bias against bursty traffic11RED 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 congestion12RED OperationMin threshMax threshAverage Queue LengthminthmaxthmaxP1.0Avg queue lengthP(drop)13Fair Queuing: Goals•How do you protect the most important packets?–How do you provide some isolation in general?–Simple priority queuing does not help•Two approaches:–Fair Queuing–Leaky bucket (with other techniques which we will cover next class)•FQ key goal: Allocate resources “fairly”–Keep separate queue for each flow•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 packet14What is “Fairness”?•At what granularity?–Flows, connections, domains?•What if users have different RTTs/links/etc.–TCP is “RTT-Fair”•BW inversely proportional to RTT of flow–Should they 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•Local notion of fairness, as we will see next–User = arbitrary granularity15Max-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 resource16Implementing 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?17Bit-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 complicated18Bit-by-bit RR Illustration•Not feasible to interleave bits on real networks–FQ simulates bit-by-bit RR19Fair 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? As we saw before, this is hard.20FQ IllustrationFlow 1Flow 2Flow nI/PO/PVariation: Weighted Fair Queuing (WFQ) Flows can have weightKey to QoS, as we will see in next class.21Bit-by-bit RR ExampleF=10Flow 1(arriving)Flow 2transmittingF=2OutputF=5F=8Flow 1 Flow 2OutputF=10Cannot preempt packetcurrently being transmitted22Fair 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
Unlocking...