An experimental study of the learnability of congestion control
|
|
|
- Amie Robinson
- 6 years ago
- Views:
Transcription
1 An experimental study of the learnability of congestion control Anirudh Sivaraman, Keith Winstein, Pratiksha Thaker, Hari Balakrishnan MIT CSAIL August 31, / 17
2 This talk How easy is it to learn a network protocol to achieve a desired goal, despite a mismatched set of assumptions? 2 / 17
3 This talk How easy is it to learn a network protocol to achieve a desired goal, despite a mismatched set of assumptions? cf. Learning: Knowledge acquisition without explicit programming (Valiant 1984) 2 / 17
4 Preview of key results 3 / 17
5 Preview of key results Can tolerate mismatched link-rate assumptions 3 / 17
6 Preview of key results Can tolerate mismatched link-rate assumptions Need precision about the number of senders 3 / 17
7 Preview of key results Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword 3 / 17
8 Preview of key results Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks 3 / 17
9 Experimental method 4 / 17
10 Experimental method 4 / 17
11 Experimental method < Mbps, ms> 4 / 17
12 Experimental method < Mbps, ms> 4 / 17
13 Experimental method < Mbps, ms> 4 / 17
14 Experimental method < Mbps, ms> 4 / 17
15 Experimental method < Mbps, ms> 4 / 17
16 Experimental method Training Networks < Mbps, ms> 5 / 17
17 Experimental method Training Networks Objective Function: - log (tpt/delay) - Avg. Flow Completion time < Mbps, ms> Learner 5 / 17
18 Experimental method Training Networks Objective Function: - log (tpt/delay) - Avg. Flow Completion time < Mbps, ms> Learner Congestion Control Algorithm 5 / 17
19 Experimental method Training Networks Objective Function: - log (tpt/delay) - Avg. Flow Completion time < Mbps, ms> Remy (SIGCOMM 13) RemyCC 5 / 17
20 Experimental method Training Networks Objective Function: - log (tpt/delay) - Avg. Flow Completion time Testing Networks < Mbps, ms> < Mbps, ms> Remy (SIGCOMM 13) RemyCC Test within ns-2 5 / 17
21 Remy compared with an ideal protocol Throughput (Mbps) Queueing delay (ms) 6 / 17
22 Remy compared with an ideal protocol 32 Ideal 16 Throughput (Mbps) Queueing delay (ms) 6 / 17
23 Remy compared with an ideal protocol Ideal RemyCC Throughput (Mbps) Queueing delay (ms) 6 / 17
24 Remy compared with an ideal protocol 32 Ideal 16 RemyCC Throughput (Mbps) Cubic Cubic/sfqCoDel Queueing delay (ms) 6 / 17
25 Learning network protocols despite mismatched assumptions 7 / 17
26 Learning network protocols despite mismatched assumptions Is there a tradeoff between operating range and generality in link rates? 7 / 17
27 Learning network protocols despite mismatched assumptions Is there a tradeoff between operating range and generality in link rates? Is there a tradeoff between performance and operating range in link rates? 7 / 17
28 Performance and link-rate operating range Objective Function (Normalized) Link rate (Mbps) 8 / 17
29 Performance and link-rate operating range 0 Ideal -0.5 Objective Function (Normalized) Link rate (Mbps) 8 / 17
30 Performance and link-rate operating range 2x range 0 Ideal -0.5 Objective Function (Normalized) Link rate (Mbps) 8 / 17
31 Performance and link-rate operating range 2x range 10x range 0 Ideal -0.5 Objective Function (Normalized) Link rate (Mbps) 8 / 17
32 Performance and link-rate operating range 0 2x range 10x range 100x range Ideal -0.5 Objective Function (Normalized) Link rate (Mbps) 8 / 17
33 Performance and link-rate operating range 0 2x range 10x range 100x range 1000x range Ideal -0.5 Objective Function (Normalized) Link rate (Mbps) 8 / 17
34 Performance and link-rate operating range 0 2x range 10x range 100x range 1000x range Ideal -0.5 Objective Function (Normalized) -1 Cubic-over-sfqCoDel -1.5 Cubic Link rate (Mbps) 8 / 17
35 Performance and link-rate operating range 9 / 17
36 Performance and link-rate operating range Very clear generality vs. operating range tradeoff 9 / 17
37 Performance and link-rate operating range Very clear generality vs. operating range tradeoff Only weak evidence of a performance vs. operating range tradeoff 9 / 17
38 Performance and link-rate operating range Very clear generality vs. operating range tradeoff Only weak evidence of a performance vs. operating range tradeoff Possible to design a forwards-comptabible protocol handling a wide range in link rates 9 / 17
39 Learning network protocols despite mismatched assumptions Can we learn a protocol that performs well both when there are few senders and when there are many senders? 10 / 17
40 Imperfections in the number of senders Normalized objective function Number of senders 11 / 17
41 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
42 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
43 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
44 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
45 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
46 Imperfections in the number of senders 0.0 Ideal 0.2 Normalized objective function Number of senders 11 / 17
47 Imperfections in the number of senders 0.0 Ideal Normalized objective function Number of senders 11 / 17
48 Imperfections in the number of senders 0.0 Ideal Normalized objective function Number of senders 11 / 17
49 Imperfections in the number of senders 0.0 Ideal Normalized objective function Number of senders 11 / 17
50 Imperfections in the number of senders 0.0 Ideal Normalized objective function Number of senders 11 / 17
51 Imperfections in the number of senders 0.0 Ideal Normalized objective function Cubic Number of senders 11 / 17
52 Imperfections in the number of senders 0.0 Ideal Normalized objective function Cubic Cubic-over-sfqCoDel Number of senders 11 / 17
53 Imperfections in the number of senders Tradeoff between performance with few senders and performance with many senders 11 / 17
54 Learning network protocols despite mismatched assumptions What are the costs and benefits of learning a new protocol that shares fairly with a legacy sender? 12 / 17
55 Imperfect assumptions about the nature of other senders TCP-Aware RemyCC: Contends with: TCP-Aware RemyCC half the time TCP NewReno half the time. 13 / 17
56 Imperfect assumptions about the nature of other senders TCP-Aware RemyCC: Contends with: TCP-Aware RemyCC half the time TCP NewReno half the time. TCP-Naive RemyCC: Contends with: TCP-Naive RemyCC all the time 13 / 17
57 RemyCC competing against itself 7 6 Throughput (Mbps) 5 NewReno RemyCC [TCP-naive] 4 Better Queueing delay (ms) 14 / 17
58 RemyCC competing against itself 7 6 Throughput (Mbps) 5 NewReno RemyCC [TCP-naive] Cost of TCP-awareness 4 Better Queueing delay (ms) 14 / 17
59 RemyCC competing against itself 7 6 Throughput (Mbps) 5 NewReno RemyCC [TCP-aware] RemyCC [TCP-naive] Cost of TCP-awareness 4 Better Queueing delay (ms) 14 / 17
60 RemyCC competing against TCP NewReno 7 NewReno 6 Throughput (Mbps) 5 4 RemyCC [TCP-naive] Better Queueing delay (ms) / 17
61 RemyCC competing against TCP NewReno 7 NewReno 6 Throughput (Mbps) 5 Effect of TCP-aware adversary 4 RemyCC [TCP-naive] Benefit of TCP-awareness Better Queueing delay (ms) 15 / 17
62 RemyCC competing against TCP NewReno 7 NewReno 6 Throughput (Mbps) 5 Effect of TCP-aware adversary NewReno RemyCC [TCP-aware] 4 RemyCC [TCP-naive] Benefit of TCP-awareness Better Queueing delay (ms) 15 / 17
63 RemyCC competing against TCP NewReno TCP awareness benefits you when needed, costs if you don t 15 / 17
64 Caveats 16 / 17
65 Caveats Remy as a proxy for an optimal learner 16 / 17
66 Caveats Remy as a proxy for an optimal learner Results may change with better learners 16 / 17
67 Caveats Remy as a proxy for an optimal learner Results may change with better learners Negative results may no longer hold 16 / 17
68 The learnability of congestion control 17 / 17
69 The learnability of congestion control Can tolerate mismatched link-rate assumptions 17 / 17
70 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders 17 / 17
71 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword 17 / 17
72 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks 17 / 17
73 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks Ongoing work in using findings: 17 / 17
74 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks Ongoing work in using findings: improve Google s datacenter transport 17 / 17
75 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks Ongoing work in using findings: improve Google s datacenter transport user-space implementation of RemyCC 17 / 17
76 The learnability of congestion control Can tolerate mismatched link-rate assumptions Need precision about the number of senders TCP compatibility is a double-edged sword Can tolerate mismatch in the # of bottlenecks Ongoing work in using findings: improve Google s datacenter transport user-space implementation of RemyCC 17 / 17
77 Backup slides 17 / 17
78 The Remy protocol synthesis procedure Protocol: range-based rule table from state to action 17 / 17
79 The Remy protocol synthesis procedure Protocol: range-based rule table from state to action State: Congestion signals tracked by the sender s ewma : EWMA over packet inter-transmit times r ewma : EWMA over ACK inter-arrival times rtt ratio: Ratio of RTT to minimum RTT slow r ewma: Slower version of s ewma 17 / 17
80 The Remy protocol synthesis procedure Protocol: range-based rule table from state to action State: Congestion signals tracked by the sender s ewma : EWMA over packet inter-transmit times r ewma : EWMA over ACK inter-arrival times rtt ratio: Ratio of RTT to minimum RTT slow r ewma: Slower version of s ewma Action: modify window, transmission rate Multiplier m to current window Increment c to current window Minimum inter-transmit time. 17 / 17
81 The Remy protocol synthesis procedure 1. Start with one rule: one action for all states 2. Optimize each action to maximize objective 3. Find most used rule 4. Median split that rule based on state usage 5. Repeat 2, 3, and 4 till you converge 17 / 17
82 One action for all states. Find the best value. r_ewma <?,?,?> s_ewma 17 / 17
83 The best (single) action. Now split it on median. r_ewma <0.90,4,3.3> s_ewma 17 / 17
84 Simulate <0.90,4,3.3> <0.90,4,3.3> r_ewma <0.90,4,3.3> <0.90,4,3.3> s_ewma 17 / 17
85 Optimize each of the new actions <0.90,4,3.3> <0.90,4,3.3> r_ewma <0.90,4,3.3> <0.90,4,3.3> s_ewma 17 / 17
86 Now split the most-used rule <0.90,5,2.8> <0.70,6,53.5> r_ewma <0.60,19,76.2> <0.80,5,4.1> s_ewma 17 / 17
87 Simulate <0.90,5,2.8> <0.70,6,53.5> r_ewma <0.80,5,4.1> <0.80,5,4.1> <0.60,19,76.2> <0.80,5,4.1> <0.80,5,4.1> s_ewma 17 / 17
88 Optimize <0.90,5,2.8> <0.70,6,53.5> r_ewma <0.80,5,4.1> <0.80,5,4.1> <0.60,19,76.2> <0.80,5,4.1> <0.80,5,4.1> s_ewma 17 / 17
89 Split <0.90,5,2.8> <0.60,17,13.3> r_ewma <0.80,8,3.3> <0.80,17,4.6> <0.30,29,49.7> <0.80,8,62.7> <0.80,7,16.9> s_ewma 17 / 17
90 Simulate <0.90,5,2.8> <0.90,5,2.8> <0.60,17,13.3> r_ewma <0.90,5,2.8> <0.90,5,2.8> <0.80,8,3.3> <0.80,17,4.6> <0.30,29,49.7> <0.80,8,62.7> <0.80,7,16.9> s_ewma 17 / 17
91 Can applications with different objectives coexist? Tpt. Sender: A throughput-intensive sender log(throughput) 0.1 log(delay) (1) Lat. Sender: A latency-sensitive sender log(throughput) 10.0 log(delay) (2) Running over a FIFO queue 17 / 17
92 Training for diversity has a cost Throughput (Mbps) Queueing delay (ms) 17 / 17
93 Training for diversity has a cost Throughput (Mbps) 5 Tpt. Sender [naive] Lat. Sender [naive] Queueing delay (ms) 17 / 17
94 Training for diversity has a cost Throughput (Mbps) Cost of Coexistence Tpt. Sender [naive] Tpt. Sender [coevolved] Lat. Sender [naive] Lat. Sender [coevolved] Queueing delay (ms) 17 / 17
95 but, benefits the docile sender Throughput (Mbps) Queueing delay (ms) 17 / 17
96 but, benefits the docile sender Tpt. Sender [naive] Throughput (Mbps) 5 2 Lat. Sender [naive] Queueing delay (ms) 17 / 17
97 but, benefits the docile sender 16 Throughput (Mbps) Tpt. Sender [naive] Effect of playing nice Lat. Sender [naive] Benefit of coevolution Tpt. Sender [coevolved] Lat. Sender [coevolved] Queueing delay (ms) 17 / 17
TCP EX MACHINA: COMPUTER-GENERATED CONGESTION CONTROL KEITH WINSTEIN AND HARI BALAKRISHNAN. Presented by: Angela Jiang
TCP EX MACHINA: COMPUTER-GENERATED CONGESTION CONTROL KEITH WINSTEIN AND HARI BALAKRISHNAN Presented by: Angela Jiang Network congestion Cause: Sources trying to send data faster than the network can process
An Experimental Study of the Learnability of Congestion Control
An Experimental Study of the Learnability of Congestion Control Anirudh Sivaraman, Keith Winstein, Pratiksha Thaker, and Hari Balakrishnan Computer Science and Artificial Intelligence Laboratory Massachusetts
PCC: Performance-oriented Congestion Control
PCC: Performance-oriented Congestion Control Michael Schapira Hebrew University of Jerusalem and Compira Labs (co-founder and chief scientist) Performance-oriented Congestion Control PCC: Re-architecting
An Experimental Study of the Learnability of Congestion Control
An Experimental Study of the Learnability of Congestion Control Anirudh Sivaraman, Keith Winstein, Pratiksha Thaker, and Hari Balakrishnan Computer Science and Artificial Intelligence Laboratory Massachusetts
Stochastic Forecasts Achieve High Throughput and Low Delay over Cellular Networks
Stochastic Forecasts Achieve High Throughput and Low Delay over Cellular Networks Keith Winstein, Anirudh Sivaraman, and Hari Balakrishnan M.I.T. CSAIL http://alfalfa.mit.edu April 5, 2013 Cellular networks
Flow-start: Faster and Less Overshoot with Paced Chirping
Flow-start: Faster and Less Overshoot with Paced Chirping Joakim Misund, Simula and Uni Oslo Bob Briscoe, Independent IRTF ICCRG, Jul 2018 The Slow-Start
Transport Layer (Congestion Control)
Transport Layer (Congestion Control) Where we are in the Course Moving on up to the Transport Layer! Application Transport Network Link Physical CSE 461 University of Washington 2 TCP to date: We can set
Congestion Control. Daniel Zappala. CS 460 Computer Networking Brigham Young University
Congestion Control Daniel Zappala CS 460 Computer Networking Brigham Young University 2/25 Congestion Control how do you send as fast as possible, without overwhelming the network? challenges the fastest
Lecture 21: Congestion Control" CSE 123: Computer Networks Alex C. Snoeren
Lecture 21: Congestion Control" CSE 123: Computer Networks Alex C. Snoeren Lecture 21 Overview" How fast should a sending host transmit data? Not to fast, not to slow, just right Should not be faster than
Performance-oriented Congestion Control
Performance-oriented Congestion Control Mo Dong, Qingxi Li, Doron Zarchy*, P. Brighten Godfrey and Michael Schapira* University of Illinois at Urbana Champaign *Hebrew University of Jerusalem Mo Dong Qingxi
6.033 Spring 2015 Lecture #11: Transport Layer Congestion Control Hari Balakrishnan Scribed by Qian Long
6.033 Spring 2015 Lecture #11: Transport Layer Congestion Control Hari Balakrishnan Scribed by Qian Long Please read Chapter 19 of the 6.02 book for background, especially on acknowledgments (ACKs), timers,
Computer Networks. Course Reference Model. Topic. Congestion What s the hold up? Nature of Congestion. Nature of Congestion 1/5/2015.
Course Reference Model Computer Networks 7 Application Provides functions needed by users Zhang, Xinyu Fall 204 4 Transport Provides end-to-end delivery 3 Network Sends packets over multiple links School
The War Between Mice and Elephants
The War Between Mice and Elephants (by Liang Guo and Ibrahim Matta) Treating Short Connections fairly against Long Connections when they compete for Bandwidth. Advanced Computer Networks CS577 Fall 2013
Internet Networking recitation #10 TCP New Reno Vs. Reno
recitation #0 TCP New Reno Vs. Reno Spring Semester 200, Dept. of Computer Science, Technion 2 Introduction Packet Loss Management TCP Reno (RFC 258) can manage a loss of at most one packet from a single
Congestion Control for High Bandwidth-delay Product Networks. Dina Katabi, Mark Handley, Charlie Rohrs
Congestion Control for High Bandwidth-delay Product Networks Dina Katabi, Mark Handley, Charlie Rohrs Outline Introduction What s wrong with TCP? Idea of Efficiency vs. Fairness XCP, what is it? Is it
Fastpass A Centralized Zero-Queue Datacenter Network
Fastpass A Centralized Zero-Queue Datacenter Network Jonathan Perry Amy Ousterhout Hari Balakrishnan Devavrat Shah Hans Fugal Ideal datacenter network properties No current design satisfies all these properties
CS519: Computer Networks. Lecture 5, Part 4: Mar 29, 2004 Transport: TCP congestion control
: Computer Networks Lecture 5, Part 4: Mar 29, 2004 Transport: TCP congestion control TCP performance We ve seen how TCP the protocol works Sequencing, receive window, connection setup and teardown And
Congestion Control in Datacenters. Ahmed Saeed
Congestion Control in Datacenters Ahmed Saeed What is a Datacenter? Tens of thousands of machines in the same building (or adjacent buildings) Hundreds of switches connecting all machines What is a Datacenter?
TCP in Asymmetric Environments
TCP in Asymmetric Environments KReSIT, IIT Bombay Vijay T. Raisinghani TCP in Asymmetric Environments 1 TCP Overview Four congestion control algorithms Slow start Congestion avoidance Fast retransmit Fast
Performance-oriented Congestion Control
Performance-oriented Congestion Control Mo Dong, Qingxi Li, Doron Zarchy*, P. Brighten Godfrey and Michael Schapira* University of Illinois at Urbana Champaign *Hebrew University of Jerusalem Qingxi Li
Impact of transmission errors on TCP performance. Outline. Random Errors
Impact of transmission errors on TCP performance 1 Outline Impact of transmission errors on TCP performance Approaches to improve TCP performance Classification Discussion of selected approaches 2 Random
Lecture 14: Congestion Control"
Lecture 14: Congestion Control" CSE 222A: Computer Communication Networks George Porter Thanks: Amin Vahdat, Dina Katabi and Alex C. Snoeren Lecture 14 Overview" TCP congestion control review Dukkipati
Packet Scheduling in Data Centers. Lecture 17, Computer Networks (198:552)
Packet Scheduling in Data Centers Lecture 17, Computer Networks (198:552) Datacenter transport Goal: Complete flows quickly / meet deadlines Short flows (e.g., query, coordination) Large flows (e.g., data
TCP congestion control:
TCP congestion control: Probing for usable bandwidth: Ideally: transmit as fast as possible (cwnd as large as possible) without loss Increase cwnd until loss (congestion) Loss: decrease cwnd, then begin
TCP Congestion Control
6.033, Spring 2014 TCP Congestion Control Dina Katabi & Sam Madden nms.csail.mit.edu/~dina Sharing the Internet How do you manage resources in a huge system like the Internet, where users with different
TCP so far Computer Networking Outline. How Was TCP Able to Evolve
TCP so far 15-441 15-441 Computer Networking 15-641 Lecture 14: TCP Performance & Future Peter Steenkiste Fall 2016 www.cs.cmu.edu/~prs/15-441-f16 Reliable byte stream protocol Connection establishments
Chapter III: Transport Layer
Chapter III: Transport Layer UG3 Computer Communications & Networks (COMN) Mahesh Marina mahesh@ed.ac.uk Slides thanks to Myungjin Lee and copyright of Kurose and Ross Principles of congestion control
Wireless TCP Performance Issues
Wireless TCP Performance Issues Issues, transport layer protocols Set up and maintain end-to-end connections Reliable end-to-end delivery of data Flow control Congestion control Udp? Assume TCP for the
Bandwidth Allocation & TCP
Bandwidth Allocation & TCP The Transport Layer Focus Application Presentation How do we share bandwidth? Session Topics Transport Network Congestion control & fairness Data Link TCP Additive Increase/Multiplicative
Outline Computer Networking. TCP slow start. TCP modeling. TCP details AIMD. Congestion Avoidance. Lecture 18 TCP Performance Peter Steenkiste
Outline 15-441 Computer Networking Lecture 18 TCP Performance Peter Steenkiste Fall 2010 www.cs.cmu.edu/~prs/15-441-f10 TCP congestion avoidance TCP slow start TCP modeling TCP details 2 AIMD Distributed,
Reliable Transport II: TCP and Congestion Control
Reliable Transport II: TCP and Congestion Control Brad Karp UCL Computer Science CS 3035/GZ01 31 st October 2013 Outline Slow Start AIMD Congestion control Throughput, loss, and RTT equation Connection
DDoS Defense by Offense
DDoS Defense by Offense Michael Walfish, Mythili Vutukuru, Hari Balakrishnan, David Karger, and Scott Shenker, SIGCOMM 06 Presented by Nikki Benecke, Nov. 7 th, 2006, for CS577 DDoS: Defense by Offense
Network Management & Monitoring Network Delay
Network Management & Monitoring Network Delay These materials are licensed under the Creative Commons Attribution-Noncommercial 3.0 Unported license (http://creativecommons.org/licenses/by-nc/3.0/) End-to-end
CS644 Advanced Networks
What we know so far CS644 Advanced Networks Lecture 6 Beyond TCP Congestion Control Andreas Terzis TCP Congestion control based on AIMD window adjustment [Jac88] Saved Internet from congestion collapse
Congestion Control In the Network
Congestion Control In the Network Brighten Godfrey cs598pbg September 9 2010 Slides courtesy Ion Stoica with adaptation by Brighten Today Fair queueing XCP Announcements Problem: no isolation between flows
One More Bit Is Enough
One More Bit Is Enough Yong Xia, RPI Lakshmi Subramanian, UCB Ion Stoica, UCB Shiv Kalyanaraman, RPI SIGCOMM 05, Philadelphia, PA 08 / 23 / 2005 Motivation #1: TCP doesn t work well in high b/w or delay
Equation-Based Congestion Control for Unicast Applications. Outline. Introduction. But don t we need TCP? TFRC Goals
Equation-Based Congestion Control for Unicast Applications Sally Floyd, Mark Handley AT&T Center for Internet Research (ACIRI) Jitendra Padhye Umass Amherst Jorg Widmer International Computer Science Institute
CSC 4900 Computer Networks: TCP
CSC 4900 Computer Networks: TCP Professor Henry Carter Fall 2017 Chapter 3 outline 3.1 Transport-layer services 3.2 Multiplexing and demultiplexing 3.3 Connectionless transport: UDP 3.4 Principles of reliable
MPTCP and BBR performance over Internet satellite paths. Nicolas KUHN
MPTCP and BBR performance over Internet satellite paths Nicolas KUHN nicolas.kuhn@cnes.fr Myth #1: SATCOM systems are quite specific Indeed: Limited frequency resource (regulation, etc.) Dish alignment
Congestion Control. Principles of Congestion Control. Network assisted congestion. Asynchronous Transfer Mode. Computer Networks 10/23/2013
Congestion Control Kai Shen Principles of Congestion Control Congestion: Informally: too many sources sending too much data too fast for the network to handle Results of congestion: long delays (e.g. queueing
Congestion Control for High Bandwidth-delay Product Networks
Congestion Control for High Bandwidth-delay Product Networks Dina Katabi, Mark Handley, Charlie Rohrs Presented by Chi-Yao Hong Adapted from slides by Dina Katabi CS598pbg Sep. 10, 2009 Trends in the Future
Topics. TCP sliding window protocol TCP PUSH flag TCP slow start Bulk data throughput
Topics TCP sliding window protocol TCP PUSH flag TCP slow start Bulk data throughput 2 Introduction In this chapter we will discuss TCP s form of flow control called a sliding window protocol It allows
Congestion Control End Hosts. CSE 561 Lecture 7, Spring David Wetherall. How fast should the sender transmit data?
Congestion Control End Hosts CSE 51 Lecture 7, Spring. David Wetherall Today s question How fast should the sender transmit data? Not tooslow Not toofast Just right Should not be faster than the receiver
Reliable Transport II: TCP and Congestion Control
Reliable Transport II: TCP and Congestion Control Stefano Vissicchio UCL Computer Science COMP0023 Recap: Last Lecture Transport Concepts Layering context Transport goals Transport mechanisms and design
THE NETWORK PERFORMANCE OVER TCP PROTOCOL USING NS2
THE NETWORK PERFORMANCE OVER TCP PROTOCOL USING NS2 Ammar Abdulateef Hadi, Raed A. Alsaqour and Syaimak Abdul Shukor School of Computer Science, Faculty of Information Science and Technology, University
The Controlled Delay (CoDel) AQM Approach to fighting bufferbloat
The Controlled Delay (CoDel) AQM Approach to fighting bufferbloat BITAG TWG Boulder, CO February 27, 2013 Kathleen Nichols Van Jacobson Background The persistently full buffer problem, now called bufferbloat,
Congestion Control 3/16/09
Congestion Control Outline Resource Allocation Queuing TCP Congestion Control Spring 009 CSE3064 Issues Two sides of the same coin pre-allocate resources so at to avoid congestion control congestion if
CS 43: Computer Networks. 19: TCP Flow and Congestion Control October 31, Nov 2, 2018
CS 43: Computer Networks 19: TCP Flow and Congestion Control October 31, Nov 2, 2018 Five-layer Internet Model Application: the application (e.g., the Web, Email) Transport: end-to-end connections, reliability
TCP Congestion Control Beyond Bandwidth-Delay Product for Mobile Cellular Networks
TCP Congestion Control Beyond Bandwidth-Delay Product for Mobile Cellular Networks Wai Kay Leong, Zixiao Wang, Ben Leong The 13th International Conference on emerging Networking EXperiments and Technologies
Congestion Control In The Internet Part 2: How it is implemented in TCP. JY Le Boudec 2014
1 Congestion Control In The Internet Part 2: How it is implemented in TCP JY Le Boudec 2014 Contents 1. Congestion control in TCP 2. The fairness of TCP 3. The loss throughput formula 4. Explicit Congestion
Computer Networking
15-441 Computer Networking Lecture 17 TCP Performance & Future Eric Anderson Fall 2013 www.cs.cmu.edu/~prs/15-441-f13 Outline TCP modeling TCP details 2 TCP Performance Can TCP saturate a link? Congestion
Operating Systems and Networks. Network Lecture 10: Congestion Control. Adrian Perrig Network Security Group ETH Zürich
Operating Systems and Networks Network Lecture 10: Congestion Control Adrian Perrig Network Security Group ETH Zürich Where we are in the Course More fun in the Transport Layer! The mystery of congestion
Where we are in the Course. Topic. Nature of Congestion. Nature of Congestion (3) Nature of Congestion (2) Operating Systems and Networks
Operating Systems and Networks Network Lecture 0: Congestion Control Adrian Perrig Network Security Group ETH Zürich Where we are in the Course More fun in the Transport Layer! The mystery of congestion
Congestion Control In The Internet Part 2: How it is implemented in TCP. JY Le Boudec 2014
1 Congestion Control In The Internet Part 2: How it is implemented in TCP JY Le Boudec 2014 Contents 1. Congestion control in TCP 2. The fairness of TCP 3. The loss throughput formula 4. Explicit Congestion
Congestion Control. Principles of Congestion Control. Network-assisted Congestion Control: ATM. Congestion Control. Computer Networks 10/21/2009
Congestion Control Kai Shen Principles of Congestion Control Congestion: informally: too many sources sending too much data too fast for the network to handle results of congestion: long delays (e.g. queueing
CMPE 257: Wireless and Mobile Networking
CMPE 257: Wireless and Mobile Networking Katia Obraczka Computer Engineering UCSC Baskin Engineering Lecture 10 CMPE 257 Spring'15 1 Student Presentations Schedule May 21: Sam and Anuj May 26: Larissa
Chapter 3 Transport Layer
Chapter 3 Transport Layer 1 Chapter 3 outline 3.1 Transport-layer services 3.2 Multiplexing and demultiplexing 3.3 Connectionless transport: UDP 3.4 Principles of reliable data transfer 3.5 Connection-oriented
CMPE 150/L : Introduction to Computer Networks. Chen Qian Computer Engineering UCSC Baskin Engineering Lecture 10
CMPE 150/L : Introduction to Computer Networks Chen Qian Computer Engineering UCSC Baskin Engineering Lecture 10 1 Midterm exam Midterm next Thursday Close book but one-side 8.5"x11" note is allowed (must
Activity-Based Congestion Management for Fair Bandwidth Sharing in Trusted Packet Networks
Communication Networks Activity-Based Congestion Management for Fair Bandwidth Sharing in Trusted Packet Networks Michael Menth and Nikolas Zeitler http://kn.inf.uni-tuebingen.de Outline The problem Definition
Problem 7. Problem 8. Problem 9
Problem 7 To best answer this question, consider why we needed sequence numbers in the first place. We saw that the sender needs sequence numbers so that the receiver can tell if a data packet is a duplicate
CSCI Topics: Internet Programming Fall 2008
CSCI 491-01 Topics: Internet Programming Fall 2008 Transport Layer Derek Leonard Hendrix College October 22, 2008 Original slides copyright 1996-2007 J.F Kurose and K.W. Ross 1 Chapter 3: Roadmap 3.1 Transport-layer
Priority Traffic CSCD 433/533. Advanced Networks Spring Lecture 21 Congestion Control and Queuing Strategies
CSCD 433/533 Priority Traffic Advanced Networks Spring 2016 Lecture 21 Congestion Control and Queuing Strategies 1 Topics Congestion Control and Resource Allocation Flows Types of Mechanisms Evaluation
Congestion control mechanism of TCP for achieving predictable throughput
Congestion control mechanism of TCP for achieving predictable throughput Kana Yamanegi Go Hasegawa Masayuki Murata Graduate School of Information Science and Technology, Osaka University 1-5 Yamadaoka,
Lecture 21. Reminders: Homework 6 due today, Programming Project 4 due on Thursday Questions? Current event: BGP router glitch on Nov.
Lecture 21 Reminders: Homework 6 due today, Programming Project 4 due on Thursday Questions? Current event: BGP router glitch on Nov. 7 http://money.cnn.com/2011/11/07/technology/juniper_internet_outage/
STMS: Improving MPTCP Throughput Under Heterogenous Networks
STMS: Improving MPTCP Throughput Under Heterogenous Networks Hang Shi 1, Yong Cui 1, Xin Wang 2, Yuming Hu 1, Minglong Dai 1, anzhao Wang 3, Kai Zheng 3 1 Tsinghua University, 2 Stony Brook University,
Lecture 15: Transport Layer Congestion Control
Lecture 15: Transport Layer Congestion Control COMP 332, Spring 2018 Victoria Manfredi Acknowledgements: materials adapted from Computer Networking: A Top Down Approach 7 th edition: 1996-2016, J.F Kurose
Improving the Robustness of TCP to Non-Congestion Events
Improving the Robustness of TCP to Non-Congestion Events Presented by : Sally Floyd floyd@acm.org For the Authors: Sumitha Bhandarkar A. L. Narasimha Reddy {sumitha,reddy}@ee.tamu.edu Problem Statement
Congestion Control in TCP
Congestion Control in TCP Antonio Carzaniga Faculty of Informatics University of Lugano May 6, 2005 Outline Intro to congestion control Input rate vs. output throughput Congestion window Congestion avoidance
Improving TCP Performance over Wireless Networks using Loss Predictors
Improving TCP Performance over Wireless Networks using Loss Predictors Fabio Martignon Dipartimento Elettronica e Informazione Politecnico di Milano P.zza L. Da Vinci 32, 20133 Milano Email: martignon@elet.polimi.it
Lecture 22: Buffering & Scheduling. CSE 123: Computer Networks Alex C. Snoeren
Lecture 22: Buffering & Scheduling CSE 123: Computer Networks Alex C. Snoeren Lecture 23 Overview Buffer Management FIFO RED Traffic Policing/Scheduling 2 Key Router Challenges Buffer management: which
CSE 123A Computer Networks
CSE 123A Computer Networks Winter 2005 Lecture 14 Congestion Control Some images courtesy David Wetherall Animations by Nick McKeown and Guido Appenzeller The bad news and the good news The bad news: new
Overview. TCP & router queuing Computer Networking. TCP details. Workloads. TCP Performance. TCP Performance. Lecture 10 TCP & Routers
Overview 15-441 Computer Networking TCP & router queuing Lecture 10 TCP & Routers TCP details Workloads Lecture 10: 09-30-2002 2 TCP Performance TCP Performance Can TCP saturate a link? Congestion control
Congestion Control In The Internet Part 2: How it is implemented in TCP. JY Le Boudec 2015
1 Congestion Control In The Internet Part 2: How it is implemented in TCP JY Le Boudec 2015 Contents 1. Congestion control in TCP 2. The fairness of TCP 3. The loss throughput formula 4. Explicit Congestion
DeTail Reducing the Tail of Flow Completion Times in Datacenter Networks. David Zats, Tathagata Das, Prashanth Mohan, Dhruba Borthakur, Randy Katz
DeTail Reducing the Tail of Flow Completion Times in Datacenter Networks David Zats, Tathagata Das, Prashanth Mohan, Dhruba Borthakur, Randy Katz 1 A Typical Facebook Page Modern pages have many components
Recap. TCP connection setup/teardown Sliding window, flow control Retransmission timeouts Fairness, max-min fairness AIMD achieves max-min fairness
Recap TCP connection setup/teardown Sliding window, flow control Retransmission timeouts Fairness, max-min fairness AIMD achieves max-min fairness 81 Feedback Signals Several possible signals, with different
EP2210 Scheduling. Lecture material:
EP2210 Scheduling Lecture material: Bertsekas, Gallager, 6.1.2. MIT OpenCourseWare, 6.829 A. Parekh, R. Gallager, A generalized Processor Sharing Approach to Flow Control - The Single Node Case, IEEE Infocom
TCP LoLa Toward Low Latency and High Throughput Congestion Control
TCP LoLa Toward Low Latency and High Throughput Congestion Control Mario Hock, Felix Neumeister, Martina Zitterbart, Roland Bless KIT The Research University in the Helmholtz Association www.kit.edu Motivation
ETSF10 Internet Protocols Transport Layer Protocols
ETSF10 Internet Protocols Transport Layer Protocols 2012, Part 2, Lecture 2.1 Kaan Bür, Jens Andersson Transport Layer Protocols Process-to-process delivery [ed.4 ch.23.1] [ed.5 ch.24.1] Transmission Control
Investigating the Use of Synchronized Clocks in TCP Congestion Control
Investigating the Use of Synchronized Clocks in TCP Congestion Control Michele Weigle Dissertation Defense May 14, 2003 Advisor: Kevin Jeffay Research Question Can the use of exact timing information improve
This Lecture: Congestion Control : Computer Networking. Introduction to TCP. Key Things You Should Know Already. Congestion Control
This Lecture: Congestion Control Congestion Control 15-744: Computer Networking L-4 TCP Assigned Reading [Chiu & Jain] Analysis of Increase and Decrease Algorithms for Congestion Avoidance in Computer
Congestion Avoidance
Congestion Avoidance Richard T. B. Ma School of Computing National University of Singapore CS 5229: Advanced Compute Networks References K. K. Ramakrishnan, Raj Jain, A Binary Feedback Scheme for Congestion
Rethinking the Timescales at Which Congestion-control Operates
Rethinking the Timescales at Which Congestion-control Operates Vishnu Konda and Jasleen Kaur University of North Carolina at Chapel Hill {vishnu, jasleen}@cs.unc.edu Abstract The efficiency of TCP congestion-control
UNIT IV -- TRANSPORT LAYER
UNIT IV -- TRANSPORT LAYER TABLE OF CONTENTS 4.1. Transport layer. 02 4.2. Reliable delivery service. 03 4.3. Congestion control. 05 4.4. Connection establishment.. 07 4.5. Flow control 09 4.6. Transmission
The Effects of Asymmetry on TCP Performance
The Effects of Asymmetry on TCP Performance Hari Balakrishnan Venkata N. Padmanabhan Randy H. Katz University of California at Berkeley Daedalus/BARWAN Retreat June 1997 Outline Overview Bandwidth asymmetry
Lecture 14: Congestion Control"
Lecture 14: Congestion Control" CSE 222A: Computer Communication Networks Alex C. Snoeren Thanks: Amin Vahdat, Dina Katabi Lecture 14 Overview" TCP congestion control review XCP Overview 2 Congestion Control
Q23-5 In a network, the size of the receive window is 1 packet. Which of the follow-ing protocols is being used by the network?
CS368: Exercise 5 Q23-5 In a network, the size of the receive window is 1 packet. Which of the follow-ing protocols is being used by the network? a) Stop_and_Wait b) Go-Back-N c) Selective-Repeat Q23-6.
Sally Floyd, Mark Handley, and Jitendra Padhye. Sept. 4-6, 2000
A Comparison of Equation-Based and AIMD Congestion Control Sally Floyd, Mark Handley, and Jitendra Padhye Sept. 4-6, 2 Workshop on the Modeling of Congestion Control Algorithms Paris 1 Why look at non-tcp
QUIC evaluation. HTTP Workshop. 28 July 2015 Münster - Germany. G. Carlucci, L. De Cicco, S. Mascolo. Politecnico di Bari, Italy
QUIC evaluation HTTP Workshop 28 July 2015 Münster - Germany G. Carlucci, L. De Cicco, S. Mascolo Politecnico di Bari, Italy Goal FOCUS OF THE TALK We want to answer to these questions: Can QUIC be safely
Evaluation of End-to-End TCP Performance over WCDMA
Evaluation of End-to-End TCP Performance over WCDMA Liang Hu Network Group, COM Centre Technical University of Denmark Outline Recall background knowledge Why study TCP over WCDMA Performance Evaluation
Chapter 3 Transport Layer
Chapter 3 Transport Layer A note on the use of these ppt slides: We re making these slides freely available to all (faculty, students, readers). They re in PowerPoint form so you can add, modify, and delete
Low Latency via Redundancy
Low Latency via Redundancy Ashish Vulimiri, Philip Brighten Godfrey, Radhika Mittal, Justine Sherry, Sylvia Ratnasamy, Scott Shenker Presenter: Meng Wang 2 Low Latency Is Important Injecting just 400 milliseconds
CSCE 463/612 Networks and Distributed Processing Spring 2017
CSCE 463/612 Networks and Distributed Processing Spring 2017 Transport Layer VI Dmitri Loguinov Texas A&M University March 28, 2017 Original slides copyright 1996-2004 J.F Kurose and K.W. Ross 1 Chapter
The War Between Mice and Elephants
The War Between Mice and Elephants Liang Guo and Ibrahim Matta Computer Science Department Boston University 9th IEEE International Conference on Network Protocols (ICNP),, Riverside, CA, November 2001.
Wireless Challenges : Computer Networking. Overview. Routing to Mobile Nodes. Lecture 24: Mobile and Wireless
Wireless Challenges 15-441: Computer Networking Lecture 24: Mobile and Wireless Peter Steenkiste Fall 2010 www.cs.cmu.edu/~prs/15-441-f10 Force us to rethink many assumptions Need to share airwaves rather
Congestion Control In The Internet Part 2: How it is implemented in TCP. JY Le Boudec 2015
Congestion Control In The Internet Part 2: How it is implemented in TCP JY Le Boudec 2015 1 Contents 1. Congestion control in TCP 2. The fairness of TCP 3. The loss throughput formula 4. Explicit Congestion
ADVANCED TOPICS FOR CONGESTION CONTROL
ADVANCED TOPICS FOR CONGESTION CONTROL Congestion Control The Internet only functions because TCP s congestion control does an effective job of matching traffic demand to available capacity. TCP s Window
RED behavior with different packet sizes
RED behavior with different packet sizes Stefaan De Cnodder, Omar Elloumi *, Kenny Pauwels Traffic and Routing Technologies project Alcatel Corporate Research Center, Francis Wellesplein, 1-18 Antwerp,
End-to-End Mechanisms for QoS Support in Wireless Networks
End-to-End Mechanisms for QoS Support in Wireless Networks R VS Torsten Braun joint work with Matthias Scheidegger, Marco Studer, Ruy de Oliveira Computer Networks and Distributed Systems Institute of
Investigating the Use of Synchronized Clocks in TCP Congestion Control
Investigating the Use of Synchronized Clocks in TCP Congestion Control Michele Weigle (UNC-CH) November 16-17, 2001 Univ. of Maryland Symposium The Problem TCP Reno congestion control reacts only to packet
MEASURING PERFORMANCE OF VARIANTS OF TCP CONGESTION CONTROL PROTOCOLS
MEASURING PERFORMANCE OF VARIANTS OF TCP CONGESTION CONTROL PROTOCOLS Harjinder Kaur CSE, GZSCCET, Dabwali Road, Bathinda, Punjab, India, sidhuharryab@gmail.com Gurpreet Singh Abstract CSE, GZSCCET, Dabwali