Unit2 Congestion Control mcsm technique to control congestion
- Slides: 23
[Unit-2] Congestion Control m/c’sm & technique to control congestion and keep load below capacity. Congestion occurs because- Routers/ Switches/ Gateways have (buffer) queues that keep or hold packets before and after processing 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 1
[Unit-2] CC Congestion in Network may occur, if LOAD on the network. . Number of Packets sent to the N/W is Greater than the Capacity of the Network, the number of packets the Network can handle. i. e. “too many sources sending too much data too fast for network to handle” may be = B’coz of lost packets (buffer overflow at routers) long delays (queueing in router buffers) 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 2
[Unit-2] CC TCP connection 1 TCP connection 2 bottleneck router capacity R When packet arrives at the interface: 1) Placed at the end of Q to get processed 2) Processing – Removes Packet – uses Routing Table & Destination Address. . . 3) Packet is placed at aprropriate outgoing Q 4). . . 5). . 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3
[Unit-2] CC Congestion involves and affects 2 Factors: 1) Delay 2) Throughput Delay = Propagation delay + Processeing delay (Negligible) As load increases. . Delay increases Throughput = # of packets/bytes passing through the network in unit time 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 4
[Unit-2] Causes/Cost of Congestion Scenarios S 1 - Two senders, One Router with infinite buffers S 2 - Two senders, One Router with finite buffers S 3 - Four senders, Four Routers with finite buffers Multiple paths 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 5
[Unit-2] Causes/Cost of Congestion Scenarios S 1 - Two senders, One Router with infinite buffers two senders, two receivers one router, infinite buffers no retransmission large delays when congested • maximum achievable throughput • • Th. Vs Load 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ Delay Vs Load 6
[Unit-2] Causes/Cost of Congestion Scenarios S 2 - Two senders, One Router with finite buffers • one router, finite buffers • sender retransmission of lost packet 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 7
[Unit-2] Causes/Cost of Congestion Scenarios S 2 - Two senders, One Router with finite buffers 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 8
[Unit-2] Causes/Cost of Congestion Scenarios S 3 - Four senders, Four Routers with finite buffers Multiple paths • four senders • multihop paths • timeout/retransmit 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 9
[Unit-2] Causes/Cost of Congestion Scenarios S 3 - Four senders, Four Routers with finite buffers Multiple paths Th. Vs Load Why? Another “cost” of congestion: when packet dropped, any “upstream transmission capacity used for that packet wasted! 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 10
[Unit-2] Approaches towards CC Two broad approaches towards congestion control: End-end congestion control: • no explicit feedback from network • congestion inferred from endsystem observed loss, delay • approach taken by TCP 10/23/2021 Network-assisted congestion control: • routers provide feedback to end systems – single bit indicating congestion (SNA, DECbit, TCP/IP ECN, ATM) – explicit rate sender should send at http: //10. 0. 0. 141/~cmmankar/ 11
[Unit-2] tcp CC 1. additive increase, multiplicative decrease 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 12
[Unit-2] tcp CC TCP Congestion Control: details • sender limits transmission: Last. Byte. Sent-Last. Byte. Acked Cong. Win • Roughly, rate = Cong. Win Bytes/sec RTT • Cong. Win is dynamic, function of perceived network congestion 10/23/2021 How does sender perceive congestion? • loss event = timeout or 3 duplicate acks • TCP sender reduces rate (Cong. Win) after loss event three mechanisms: – AIMD – slow start – conservative after timeout events http: //10. 0. 0. 141/~cmmankar/ 3 -13
[Unit-2] tcp CC TCP Slow Start • When connection begins, Cong. Win = 1 MSS – Example: MSS = 500 bytes & RTT = 200 msec – initial rate = 20 kbps r When connection begins, increase rate exponentially fast until first loss event • available bandwidth may be >> MSS/RTT – desirable to quickly ramp up to respectable rate 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -14
[Unit-2] tcp CC TCP Slow Start (more) – double Cong. Win every RTT – done by incrementing Cong. Win for every ACK received Host A Host B one segm ent RTT • When connection begins, increase rate exponentially until first loss event: • Summary: initial rate is slow but ramps up exponentially fast two segm ents four segm ents time 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -15
[Unit-2] tcp CC Refinement Q: When should the exponential increase switch to linear? A: When Cong. Win gets to 1/2 of its value before timeout. Implementation: • Variable Threshold • At loss event, Threshold is set to 1/2 of Cong. Win just before loss event 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -16
[Unit-2] tcp CC Refinement: inferring loss • • After 3 dup ACKs: – Cong. Win is cut in half – window then grows linearly But after timeout event: – Cong. Win instead set to 1 MSS; – window then grows exponentially – to a threshold, then grows linearly 10/23/2021 Philosophy: q 3 dup ACKs indicates network capable of delivering some segments q timeout indicates a “more alarming” congestion scenario http: //10. 0. 0. 141/~cmmankar/ 3 -17
[Unit-2] tcp CC Summary: TCP Congestion Control • When Cong. Win is below Threshold, sender in slowstart phase, window grows exponentially. • When Cong. Win is above Threshold, sender is in congestion-avoidance phase, window grows linearly. • When a triple duplicate ACK occurs, Threshold set to Cong. Win/2 and Cong. Win set to Threshold. • When timeout occurs, Threshold set to Cong. Win/2 and Cong. Win is set to 1 MSS. 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -18
[Unit-2] tcp CC TCP sender congestion control State Event TCP Sender Action Commentary Slow Start (SS) ACK receipt Cong. Win = Cong. Win + MSS, for previously If (Cong. Win > Threshold) unacked data set state to “Congestion Avoidance” Resulting in a doubling of Cong. Win every RTT Congestion Avoidance (CA) ACK receipt Cong. Win = Cong. Win+MSS * for previously (MSS/Cong. Win) unacked data Additive increase, resulting in increase of Cong. Win by 1 MSS every RTT SS or CA Loss event detected by triple duplicate ACK Threshold = Cong. Win/2, Cong. Win = Threshold, Set state to “Congestion Avoidance” Fast recovery, implementing multiplicative decrease. Cong. Win will not drop below 1 MSS. SS or CA Timeout Threshold = Cong. Win/2, Cong. Win = 1 MSS, Set state to “Slow Start” Enter slow start SS or CA Duplicate ACK Increment duplicate ACK count for segment being acked Cong. Win and Threshold not changed 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -19
[Unit-2] tcp CC TCP throughput • What’s the average throughout of TCP as a function of window size and RTT? – Ignore slow start • Let W be the window size when loss occurs. • When window is W, throughput is W/RTT • Just after loss, window drops to W/2, throughput to W/2 RTT. • Average throughout: . 75 W/RTT 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -20
[Unit-2] tcp CC TCP Futures • Example: 1500 byte segments, 100 ms RTT, want 10 Gbps throughput • Requires window size W = 83, 333 in-flight segments 10/23/2021 http: //10. 0. 0. 141/~cmmankar/ 3 -21
[Unit-2] tcp CC TCP Fairness goal: if K TCP sessions share same bottleneck link of bandwidth R, each should have average rate of R/K TCP connection 1 TCP connection 2 10/23/2021 bottleneck router capacity R http: //10. 0. 0. 141/~cmmankar/ 3 -22
[Unit-2] tcp CC Why is TCP fair? Two competing sessions: • Additive increase gives slope of 1, as throughout increases • multiplicative decreases throughput proportionally Connection 2 throughput R equal bandwidth share loss: decrease window by factor of 2 congestion avoidance: additive increase Connection 1 throughput 10/23/2021 R http: //10. 0. 0. 141/~cmmankar/ 3 -23
- Pincuegula
- Principles of congestion control
- In2140
- Tcp congestion control
- Traffic throttling and load shedding
- What are the general principles of congestion control
- Principles of congestion control in computer networks
- Congestion control in virtual circuit
- Udp congestion control
- New reno tcp
- Principles of congestion control
- Congestion control in network layer
- Choke packets in computer networks
- Tcp header length
- General principles of congestion control
- Network congestion causes
- Traffic congestion although abu dhabi
- Traffic congestion conclusion
- Cvc lung
- Spleen congestion
- Acute pulmonary congestion histology
- Dr mary bove
- Heart failure cells are seen in lungs
- Congestion