Summary: TCP Congestion Control. When CongWin is below Threshold, sender in slowstart phase, window grows exponentially.

Transcription

1 ummary: TCP Congestion Control When CongWin is below Threshold, sender in slowstart phase, window grows exponentially. When CongWin is above Threshold, sender is in congestion-avoidance phase, window grows linearly. When a triple duplicate ACK occurs, Threshold set to CongWin/2 and CongWin set to Threshold. When timeout occurs, Threshold set to CongWin/2 and CongWin is set to M. Transport Layer 3- TCP Fairness Fairness goal: if K TCP sessions share same bottleneck link of bandwidth, each should have average rate of /K TCP connection TCP connection 2 bottleneck router capacity Transport Layer 3-2 Transport Layer IV

2 Why is TCP fair? Two competing sessions: Additive increase gives slope of, as throughout increases multiplicative decrease decreases throughput proportionally Connection 2 throughput Connection throughput equal bandwidth share loss: decrease window by factor of 2 congestion avoidance: additive increase loss: decrease window by factor of 2 congestion avoidance: additive increase Transport Layer 3-3 Fairness (more) Fairness and UDP Multimedia apps often do not use TCP do not want rate throttled by congestion control Instead use UDP: pump audio/video at constant rate, tolerate packet loss esearch area: TCP friendly Fairness and parallel TCP connections nothing prevents app from opening parallel cnctions between 2 hosts. Web browsers do this Example: link of rate supporting 9 cnctions; new app asks for TCP, gets rate /0 new app asks for TCPs, gets /2! Transport Layer 3-4 Transport Layer IV 2

3 Delay modeling Q: How long does it take to receive an object from a Web server after sending a request? Ignoring congestion, delay is influenced by: TCP connection establishment data transmission delay slow start Notation, assumptions: Assume one link between client and server of rate : M (bits) O: object size (bits) no retransmissions (no loss, no corruption) Window size: First assume: fixed congestion window, W segments Then dynamic window, modeling slow start Transport Layer 3-5 Fixed congestion window () First case: W/ > TT + /: ACK for first segment in window returns before window s worth of data sent delay = 2TT + O/ Transport Layer 3-6 Transport Layer IV 3

4 Fixed congestion window (2) econd case: W/ < TT + /: wait for ACK after sending window s worth of data sent delay = 2TT + O/ + (K-)[/ + TT - W/] Transport Layer 3-7 TCP Delay Modeling: low tart () Now suppose window grows according to slow start Will show that the delay for one object is: O P Latency = 2TT + + P TT (2 ) + where P is the number of times TCP idles at server: P = min{ Q, K } - where Q is the number of times the server idles if the object were of infinite size. - and K is the number of windows that cover the object. Transport Layer 3-8 Transport Layer IV 4

5 TCP Delay Modeling: low tart (2) Delay components: 2 TT for connection estab and request O/ to transmit object time server idles due to slow start erver idles: P = min{k-,q} times initiate TCP connection request object TT first window = / second window = 2/ third window = 4/ Example: O/ = 5 segments K = 4 windows Q = 2 P = min{k-,q} = 2 erver idles P=2 times object delivered time at client time at server fourth window = 8/ complete transmission Transport Layer TT TCP Delay Modeling (3) = time from when server starts to send until server receives acknowledg ement segment k 2 = time to transmit the kth window + k 2 + TT = idle time after the kth window initiate TCP connection request object TT first window = / second window = 2/ third window = 4/ O delay = + 2TT + P p= idletime P O k = + 2TT + [ + TT 2 k = O = + 2TT + P[ TT + ] (2 p P ] ) object delivered time at client time at server fourth window = 8/ complete transmission Transport Layer 3-0 Transport Layer IV 5

6 TCP Delay Modeling (4) ecall K = number of windows that cover object How do we calculate K? K = min{ k : 2 = min{ k : L L+ 2 k k O = min{ k : 2 } O = min{ k : k log2( + )} O = log2( + ) k O} O / } Calculation of Q, number of idles for infinite-size object, is similar (see HW). Transport Layer 3- HTTP Modeling Assume Web page consists of: base HTML page (of size O bits) M images (each of size O bits) Non-persistent HTTP: M+ TCP connections in series esponse time = (M+)O/ + (M+)2TT + sum of idle times Persistent HTTP: 2 TT to request and receive base HTML file TT to request and receive M images esponse time = (M+)O/ + 3TT + sum of idle times Non-persistent HTTP with X parallel connections uppose M/X integer. TCP connection for base file M/X sets of parallel connections for images. esponse time = (M+)O/ + (M/X + )2TT + sum of idle times Transport Layer 3-2 Transport Layer IV 6

7 HTTP esponse time (in seconds) TT = 00 msec, O = 5 Kbytes, M=0 and X= non-persistent persistent parallel nonpersistent 28 Kbps 00 Kbps Mbps 0 Mbps For low bandwidth, connection & response time dominated by transmission time. Persistent connections only give minor improvement over parallel connections. Transport Layer 3-3 HTTP esponse time (in seconds) TT = sec, O = 5 Kbytes, M=0 and X= Kbps 00 Kbps Mbps 0 Mbps non-persistent persistent parallel nonpersistent For larger TT, response time dominated by TCP establishment & slow start delays. Persistent connections now give important improvement: particularly in high delay bandwidth networks. Transport Layer 3-4 Transport Layer IV 7

8 Chapter 3: ummary principles behind transport layer services: multiplexing, demultiplexing reliable data transfer flow control congestion control instantiation and implementation in the Internet UDP TCP Next: leaving the network edge (application, transport layers) into the network core Transport Layer 3-5 Transport Layer IV 8