TCP Pacing in Data Center Networks
|
|
|
- Earl Paul Harmon
- 10 years ago
- Views:
Transcription
1 TCP Pacing in Data Center Networks Monia Ghobadi, Yashar Ganjali Department of Computer Science, University of Toronto {monia, 1
2 TCP, Oh TCP! 2
3 TCP, Oh TCP! TCP congestion control 2
4 TCP, Oh TCP! TCP congestion control Focus on evolution of cwnd over RTT. 2
5 TCP, Oh TCP! TCP congestion control Focus on evolution of cwnd over RTT. Damages 2
6 TCP, Oh TCP! TCP congestion control Focus on evolution of cwnd over RTT. Damages TCP pacing 2
7 The Tortoise and the Hare: Why Bother Pacing? 3
8 The Tortoise and the Hare: Why Bother Pacing? Renewed interest in pacing for the data center environment Small buffer switches Small round-trip times Disparity between the total capacity of the network and the capacity of individual queues 3
9 The Tortoise and the Hare: Why Bother Pacing? Renewed interest in pacing for the data center environment Small buffer switches Small round-trip times Disparity between the total capacity of the network and the capacity of individual queues Focus on tail latency cause by short-term unfairness in TCP 3
10 TCP Pacing s Potential 4
11 TCP Pacing s Potential Better link utilization on small switch buffers 4
12 TCP Pacing s Potential Better link utilization on small switch buffers Better short-term fairness among flows of similar RTTs: Improves worst-flow latency 4
13 TCP Pacing s Potential Better link utilization on small switch buffers Better short-term fairness among flows of similar RTTs: Improves worst-flow latency Allows slow-start to be circumvented Saving many round-trip time May allow much larger initial congestion window to be used safely 4
14 Contributions 5
15 Contributions Effectiveness of TCP pacing in data centers. 5
16 Contributions Effectiveness of TCP pacing in data centers. Benefits of using TCP diminish as we increase the number of concurrent connections beyond a certain threshold (Point of Inflection). 5
17 Contributions Effectiveness of TCP pacing in data centers. Benefits of using TCP diminish as we increase the number of concurrent connections beyond a certain threshold (Point of Inflection). Inconclusive results in previous works. 5
18 Contributions Effectiveness of TCP pacing in data centers. Benefits of using TCP diminish as we increase the number of concurrent connections beyond a certain threshold (Point of Inflection). Inconclusive results in previous works. Inter-flow bursts. 5
19 Contributions Effectiveness of TCP pacing in data centers. Benefits of using TCP diminish as we increase the number of concurrent connections beyond a certain threshold (Point of Inflection). Inconclusive results in previous works. Inter-flow bursts. Test-bed experiments. 5
20 Inter-flow Bursts 6
21 Inter-flow Bursts C: bottleneck link capacity 6
22 Inter-flow Bursts C: bottleneck link capacity B max :buffer size 6
23 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. 6
24 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. 6
25 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. RTT 2RTT 3RTT 6
26 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. RTT 2RTT 3RTT 6
27 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. RTT 2RTT 3RTT 6
28 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. RTT 2RTT 3RTT 6
29 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. X: Inter-flow burst RTT 2RTT 3RTT 6
30 Inter-flow Bursts C: bottleneck link capacity B max :buffer size N: longed lived flows. W: packets in every RTT in or non- manner. X: Inter-flow burst RTT 2RTT 3RTT 6
31 Modeling 7
32 Modeling best case of non- 7
33 Modeling best case of non- worst case of 7
34 Modeling best case of non- worst case of RTT 7
35 Modeling best case of non- worst case of RTT RTT 7
36 Modeling best case of non- worst case of RTT RTT 7
37 Modeling best case of non- worst case of RTT RTT 7
38 Experimental Studies 8
39 Experimental Studies Flow of sizes 1,2, 3 MB between servers and clients. 8
40 Experimental Studies Flow of sizes 1,2, 3 MB between servers and clients. Bottleneck BW: 1,2, 3 Gbps 8
41 Experimental Studies Flow of sizes 1,2, 3 MB between servers and clients. Bottleneck BW: 1,2, 3 Gbps RTT: 1 to 1ms 8
42 Experimental Studies Flow of sizes 1,2, 3 MB between servers and clients. Bottleneck BW: 1,2, 3 Gbps RTT: 1 to 1ms Bottleneck utilization, Drop rate, average and tail FCT 8
43 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT 9
44 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT No congestion 9
45 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT No congestion Congestion 9
46 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion Congestion Time (sec) 9
47 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion Congestion Time (sec) 9
48 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) 9
49 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) 9
50 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) 9
51 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) CDF Flow Completion Time (sec) 9
52 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) CDF RTT Flow Completion Time (sec) 9
53 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) CDF RTT 2RTTs Flow Completion Time (sec) 9
54 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) CDF RTT 2RTTs Flow Completion Time (sec) CDF Flow Completion Time (sec) 9
55 Base-Case Experiment: One flow vs Two flows, 64KB of buffering, Utilization/Drop/FCT Bottleneck Link Utilization (Mbps) No congestion 38% Congestion Time (sec) CDF RTT 2RTTs Flow Completion Time (sec) CDF RTTs Flow Completion Time (sec) 9
56 Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows 1
57 Bottleneck Link Utilization (Mbps) Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows N * PoI Number of Flows 1
58 Bottleneck Link Utilization (Mbps) Bottleneck Link Drop (%) Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows N * PoI Number of Flows N * PoI Number of Flows 1
59 Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) Bottleneck Link Drop (%) Number of Flows N * PoI N * PoI Average FCT (sec) N * PoI Number of Flows Number of Flows 1
60 Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) Bottleneck Link Drop (%) Number of Flows N * PoI N * PoI Number of Flows Average FCT (sec) 99 th Percentile FCT (sec) Number of Flows N * PoI N * PoI Number of Flows 1
61 Multiple flows: Link Utilization/Drop/Latency Buffer size 1.7% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) Bottleneck Link Drop (%) Number of Flows N * PoI N * PoI Number of Flows Average FCT (sec) 99 th Percentile FCT (sec) Number of Flows N * PoI Number of concurrent connections increase beyond a certain point the benefits of pacing diminish. N * PoI Number of Flows 1
62 Multiple flows: Link Utilization/Drop/Latency Buffer size 3.4% of BDP, varying number of flows 11
63 Multiple flows: Link Utilization/Drop/Latency Buffer size 3.4% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) 1 Bottleneck link drop(%) Number of Flows non N * PoI Number of flows sharing the bottleneck Average FCT (sec) 99 th Percentile FCT (sec) N * PoI Number 2 N * of Flows PoI Number of Flows 11
64 Multiple flows: Link Utilization/Drop/Latency Buffer size 3.4% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) 1 Bottleneck link drop(%) flows, BDP 125.4packets and buffer size packets N* = 8 flows. Number of Flows non N * PoI Number of flows sharing the bottleneck Average FCT (sec) Aagarwal et al.: Don t pace! Kulik et al.: Pace! 99 th Percentile FCT (sec) N * PoI Number 2 N * of Flows PoI Number of Flows 1 flow, BDP 91 packets, buffer size 1 packets. N* = 9 flows. 11
65 N* vs. Buffer 12
66 N* vs. Buffer Bottleneck Link Utilization (Mbps) Buffer Size (KB) 12
67 N* vs. Buffer Bottleneck Link Utilization (Mbps) Buffer Size (KB) non Bottleneck link drop(%) Buffer size(kb) 12
68 N* vs. Buffer Bottleneck Link Utilization (Mbps) Buffer Size (KB) Average CT (sec) Buffer Size (KB).8.7 non.6 Bottleneck link drop(%) Buffer size(kb) 12
69 N* vs. Buffer Bottleneck Link Utilization (Mbps) Bottleneck link drop(%) Buffer Size (KB) non Buffer size(kb) Average CT (sec) 99 th Percentile CT (sec) Buffer Size (KB) Buffer Size (KB) 12
70 N* vs. Buffer Bottleneck Link Utilization (Mbps) Bottleneck link drop(%) Buffer Size (KB) non Buffer size(kb) Average CT (sec) 99 th Percentile CT (sec) Buffer Size (KB) Buffer Size (KB) 12
71 Clustering Effect: The probability of packets from a flow being followed by packets from other flows 13
72 Clustering Effect: The probability of packets from a flow being followed by packets from other flows Non-: Packets of each flow are clustered together. 13
73 Clustering Effect: The probability of packets from a flow being followed by packets from other flows Non-: Packets of each flow are clustered together. Paced: Packets of different flows are multiplexed. 13
74 Drop Synchronization: Number of Flows Affected by Drop Event 14
75 Drop Synchronization: Number of Flows Affected by Drop Event NetFPGA router to count the number of flows affected by drop events. 14
76 Drop Synchronization: Number of Flows Affected by Drop Event NetFPGA router to count the number of flows affected by drop events. CDF Number of Flows Affected by Drop Event CDF Number of Flows Affected by Drop Event CDF Number of Flows Affected by Drop Event N: 48 N: 96 N:
77 Drop Synchronization: Number of Flows Affected by Drop Event NetFPGA router to count the number of flows affected by drop events. CDF Number of Flows Affected by Drop Event CDF Number of Flows Affected by Drop Event CDF Number of Flows Affected by Drop Event N: 48 N: 96 N:
78 Future Trends for Pacing: per-egress pacing. 15
79 Future Trends for Pacing: per-egress pacing. Bottleneck Link Utilization (Mbps) 1 Bottleneck Link Drop (%) per flow per host + per flow Number of Flows N * PoI per flow non per host + per flow Number of flows Average RCT (sec) 99 th Percentile RCT (sec) N * PoI per flow per host + per flow Number of Flows N * PoI per flow per host + per flow Number of Flows
80 Future Trends for Pacing: per-egress pacing. Bottleneck Link Utilization (Mbps) 1 Bottleneck Link Drop (%) per flow per host + per flow Number of Flows N * PoI per flow non per host + per flow Number of flows Average RCT (sec) 99 th Percentile RCT (sec) N * PoI per flow per host + per flow Number of Flows N * PoI per flow per host + per flow Number of Flows
81 Future Trends for Pacing: per-egress pacing. Bottleneck Link Utilization (Mbps) 1 Bottleneck Link Drop (%) per flow per host + per flow Number of Flows N * PoI per flow non per host + per flow Number of flows Average RCT (sec) 99 th Percentile RCT (sec) N * PoI per flow per host + per flow Number of Flows N * PoI per flow per host + per flow Number of Flows
82 Future Trends for Pacing: per-egress pacing. Bottleneck Link Utilization (Mbps) 1 Bottleneck Link Drop (%) per flow per host + per flow Number of Flows N * PoI per flow non per host + per flow Number of flows Average RCT (sec) 99 th Percentile RCT (sec) N * PoI per flow per host + per flow Number of Flows N * PoI per flow per host + per flow Number of Flows
83 Future Trends for Pacing: per-egress pacing. Bottleneck Link Utilization (Mbps) 1 Bottleneck Link Drop (%) per flow per host + per flow Number of Flows N * PoI per flow non per host + per flow Number of flows Average RCT (sec) 99 th Percentile RCT (sec) N * PoI per flow per host + per flow Number of Flows N * PoI per flow per host + per flow Number of Flows
84 Conclusions and Future work Re-examine TCP pacing s effectiveness: Demonstrate when TCP pacing brings benefits in such environments. Inter-flow burstiness Burst-pacing vs. packet-pacing. Per-egress pacing. 16
85 Renewed Interest 17
86 Traffic Burstiness Survey Bursty is a word with no agreed meaning. How do you define a bursty traffic? If you are involved with a data center, is your data center traffic bursty? If yes, do you think that it will be useful to supress the burstiness in your traffic? If no, are you already supressing the burstiness? How? Would you anticipate the traffic becoming burstier in the future? [email protected] 18
87 19
88 Base-Case Experiment: One RPC vs Two RPCs, 64KB of buffering, Latency 2
89 Multiple flows: Link Utilization/Drop/ Latency Buffer size: 6% of BDP, varying number of 21
90 Base-Case Experiment: One RPC vs Two RPCs, 64KB of buffering, Latency / Queue Occupancy 22
91 Base-Case Experiment: One RPC vs Two RPCs, 64KB of buffering, Latency / Queue Occupancy 22
92 Base-Case Experiment: One RPC vs Two RPCs, 64KB of buffering, Latency / Queue Occupancy 22
93 Base-Case Experiment: One RPC vs Two RPCs, 64KB of buffering, Latency / Queue Occupancy 22
94 Functional test 23
95 Functional test 23
96 Functional test 23
97 Functional test 23
98 RPC vs. Streaming Paced by ack clocking Bursty 1GE 1GE 1GE RTT = 1ms 24
99 Zooming in more on the flow 25
100 Multiple flows: Link Utilization/Drop/Latency Buffer size 6.8% of BDP, varying number of flows 26
101 Multiple flows: Link Utilization/Drop/Latency Buffer size 6.8% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) N * PoI Number of Flows 26
102 Multiple flows: Link Utilization/Drop/Latency Buffer size 6.8% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) N * PoI Number of Flows 1.8 non.6 Bottleneck link drop(%) Number of flows sharing the bottleneck 26
103 Multiple flows: Link Utilization/Drop/Latency Buffer size 6.8% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) N * PoI Number of Flows Average FCT (sec) N * PoI Number of Flows 1.8 non.6 Bottleneck link drop(%) Number of flows sharing the bottleneck 26
104 Multiple flows: Link Utilization/Drop/Latency Buffer size 6.8% of BDP, varying number of flows Bottleneck Link Utilization (Mbps) Bottleneck link drop(%) Number of Flows N * PoI non Number of flows sharing the bottleneck Average FCT (sec) 99 th Percentile FCT (sec) Number of Flows N * N * PoI PoI Number of Flows 26
TCP Pacing in Data Center Networks
TCP Pacing in Data Center Networks Monia Ghobadi Department of Computer Science University of Toronto Email: [email protected] Yashar Ganjali Department of Computer Science University of Toronto Email:
17: Queue Management. Queuing. Mark Handley
17: Queue Management Mark Handley Queuing The primary purpose of a queue in an IP router is to smooth out bursty arrivals, so that the network utilization can be high. But queues add delay and cause jitter.
Lecture 15: Congestion Control. CSE 123: Computer Networks Stefan Savage
Lecture 15: Congestion Control CSE 123: Computer Networks Stefan Savage Overview Yesterday: TCP & UDP overview Connection setup Flow control: resource exhaustion at end node Today: Congestion control Resource
Data Networks Summer 2007 Homework #3
Data Networks Summer Homework # Assigned June 8, Due June in class Name: Email: Student ID: Problem Total Points Problem ( points) Host A is transferring a file of size L to host B using a TCP connection.
TCP Adaptation for MPI on Long-and-Fat Networks
TCP Adaptation for MPI on Long-and-Fat Networks Motohiko Matsuda, Tomohiro Kudoh Yuetsu Kodama, Ryousei Takano Grid Technology Research Center Yutaka Ishikawa The University of Tokyo Outline Background
Data Center TCP (DCTCP)
Data Center TCP () Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitendra Padhye, Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan Microsoft Research Stanford University {albert,
ICTCP: Incast Congestion Control for TCP in Data Center Networks
ICTCP: Incast Congestion Control for TCP in Data Center Networks Haitao Wu, Zhenqian Feng, Chuanxiong Guo, Yongguang Zhang {hwu, v-zhfe, chguo, ygz}@microsoft.com, Microsoft Research Asia, China School
Application Level Congestion Control Enhancements in High BDP Networks. Anupama Sundaresan
Application Level Congestion Control Enhancements in High BDP Networks Anupama Sundaresan Organization Introduction Motivation Implementation Experiments and Results Conclusions 2 Developing a Grid service
Load Balancing in Data Center Networks
Load Balancing in Data Center Networks Henry Xu Computer Science City University of Hong Kong HKUST, March 2, 2015 Background Aggregator Aggregator Aggregator Worker Worker Worker Worker Low latency for
CS551 End-to-End Internet Packet Dynamics [Paxson99b]
![CS551 End-to-End Internet Packet Dynamics [Paxson99b]](/thumbs/27/11353987.jpg "CS551 End-to-End Internet Packet Dynamics [Paxson99b]")
CS551 End-to-End Internet Packet Dynamics [Paxson99b] Bill Cheng http://merlot.usc.edu/cs551-f12 1 End-to-end Packet Dynamics How do you measure Internet performance? Why do people want to know? Are ISPs
High-Speed TCP Performance Characterization under Various Operating Systems
High-Speed TCP Performance Characterization under Various Operating Systems Y. Iwanaga, K. Kumazoe, D. Cavendish, M.Tsuru and Y. Oie Kyushu Institute of Technology 68-4, Kawazu, Iizuka-shi, Fukuoka, 82-852,
How Router Technology Shapes Inter-Cloud Computing Service Architecture for The Future Internet
How Router Technology Shapes Inter-Cloud Computing Service Architecture for The Future Internet Professor Jiann-Liang Chen Friday, September 23, 2011 Wireless Networks and Evolutional Communications Laboratory
4 High-speed Transmission and Interoperability
4 High-speed Transmission and Interoperability Technology 4-1 Transport Protocols for Fast Long-Distance Networks: Comparison of Their Performances in JGN KUMAZOE Kazumi, KOUYAMA Katsushi, HORI Yoshiaki,
Congestion Control Review. 15-441 Computer Networking. Resource Management Approaches. Traffic and Resource Management. What is congestion control?
Congestion Control Review What is congestion control? 15-441 Computer Networking What is the principle of TCP? Lecture 22 Queue Management and QoS 2 Traffic and Resource Management Resource Management
Improving our Evaluation of Transport Protocols. Sally Floyd Hamilton Institute July 29, 2005
Improving our Evaluation of Transport Protocols Sally Floyd Hamilton Institute July 29, 2005 Computer System Performance Modeling and Durable Nonsense A disconcertingly large portion of the literature
MMPTCP: A Novel Transport Protocol for Data Centre Networks
MMPTCP: A Novel Transport Protocol for Data Centre Networks Morteza Kheirkhah FoSS, Department of Informatics, University of Sussex Modern Data Centre Networks FatTree It provides full bisection bandwidth
Trickle: Rate Limiting YouTube Video Streaming
Trickle: Rate Limiting YouTube Video Streaming Monia Ghobadi University of Toronto [email protected] Yuchung Cheng Ankur Jain Matt Mathis Google {ycheng, jankur, mattmathis}@google.com Abstract YouTube
Effects of Filler Traffic In IP Networks. Adam Feldman April 5, 2001 Master s Project
Effects of Filler Traffic In IP Networks Adam Feldman April 5, 2001 Master s Project Abstract On the Internet, there is a well-documented requirement that much more bandwidth be available than is used
Traffic Mangement in ATM Networks Dollar Day Sale
Traffic Mangement in ATM Networks Dollar Day Sale One Megabit memory, One Megabyte disk, One Mbps link, One MIP processor, 10 cents each... Professor of Computer and Information Sciences Columbus, OH 43210-1277
Using TrueSpeed VNF to Test TCP Throughput in a Call Center Environment
Using TrueSpeed VNF to Test TCP Throughput in a Call Center Environment TrueSpeed VNF provides network operators and enterprise users with repeatable, standards-based testing to resolve complaints about
Strategies. Addressing and Routing
Strategies Circuit switching: carry bit streams original telephone network Packet switching: store-and-forward messages Internet Spring 2007 CSE 30264 14 Addressing and Routing Address: byte-string that
Question: 3 When using Application Intelligence, Server Time may be defined as.
1 Network General - 1T6-521 Application Performance Analysis and Troubleshooting Question: 1 One component in an application turn is. A. Server response time B. Network process time C. Application response
Delay-Based Early Congestion Detection and Adaptation in TCP: Impact on web performance
1 Delay-Based Early Congestion Detection and Adaptation in TCP: Impact on web performance Michele C. Weigle Clemson University Clemson, SC 29634-196 Email: [email protected] Kevin Jeffay and F. Donelson
Analysis of Internet Transport Service Performance with Active Queue Management in a QoS-enabled Network
University of Helsinki - Department of Computer Science Analysis of Internet Transport Service Performance with Active Queue Management in a QoS-enabled Network Oriana Riva [email protected] Contents
First Midterm for ECE374 02/25/15 Solution!!
1 First Midterm for ECE374 02/25/15 Solution!! Instructions: Put your name and student number on each sheet of paper! The exam is closed book. You have 90 minutes to complete the exam. Be a smart exam
TCP/IP Over Lossy Links - TCP SACK without Congestion Control
Wireless Random Packet Networking, Part II: TCP/IP Over Lossy Links - TCP SACK without Congestion Control Roland Kempter The University of Alberta, June 17 th, 2004 Department of Electrical And Computer
Iperf Tutorial. Jon Dugan <[email protected]> Summer JointTechs 2010, Columbus, OH
Iperf Tutorial Jon Dugan Summer JointTechs 2010, Columbus, OH Outline What are we measuring? TCP Measurements UDP Measurements Useful tricks Iperf Development What are we measuring? Throughput?
Low-rate TCP-targeted Denial of Service Attack Defense
Low-rate TCP-targeted Denial of Service Attack Defense Johnny Tsao Petros Efstathopoulos University of California, Los Angeles, Computer Science Department Los Angeles, CA E-mail: {johnny5t, pefstath}@cs.ucla.edu
Basic Multiplexing models. Computer Networks - Vassilis Tsaoussidis
Basic Multiplexing models? Supermarket?? Computer Networks - Vassilis Tsaoussidis Schedule Where does statistical multiplexing differ from TDM and FDM Why are buffers necessary - what is their tradeoff,
TCP Tuning Techniques for High-Speed Wide-Area Networks. Wizard Gap
NFNN2, 20th-21st June 2005 National e-science Centre, Edinburgh TCP Tuning Techniques for High-Speed Wide-Area Networks Distributed Systems Department Lawrence Berkeley National Laboratory http://gridmon.dl.ac.uk/nfnn/
PINK: Proactive INjection into ack, a queue manager to impose fair resource allocation among TCP ows
PINK: Proactive INjection into ack, a queue manager to impose fair resource allocation among TCP ows Carlo A. Grazia, Martin Klapez, Natale Patriciello, Maurizio Casoni Department of Engineering Enzo Ferrari
Denial-of-Service Shrew Attacks
Denial-of-Service Shrew Attacks Bhuvana Mahalingam [email protected] 1. Introduction A Denial of Service Attack is defined as An incident in which a user or organization is deprived of the services of
TCP in Wireless Mobile Networks
TCP in Wireless Mobile Networks 1 Outline Introduction to transport layer Introduction to TCP (Internet) congestion control Congestion control in wireless networks 2 Transport Layer v.s. Network Layer
Requirements for Simulation and Modeling Tools. Sally Floyd NSF Workshop August 2005
Requirements for Simulation and Modeling Tools Sally Floyd NSF Workshop August 2005 Outline for talk: Requested topic: the requirements for simulation and modeling tools that allow one to study, design,
Using Linux Traffic Control on Virtual Circuits J. Zurawski Internet2 [email protected] February 25 nd 2013
Using Linux Traffic Control on Virtual Circuits J. Zurawski Internet2 [email protected] February 25 nd 2013 1. Abstract Research and Education (R&E) networks have experimented with the concept of
15-441: Computer Networks Homework 2 Solution
5-44: omputer Networks Homework 2 Solution Assigned: September 25, 2002. Due: October 7, 2002 in class. In this homework you will test your understanding of the TP concepts taught in class including flow
Netest: A Tool to Measure the Maximum Burst Size, Available Bandwidth and Achievable Throughput
Netest: A Tool to Measure the Maximum Burst Size, Available Bandwidth and Achievable Throughput Guojun Jin Brian Tierney Distributed Systems Department Lawrence Berkeley National Laboratory 1 Cyclotron
Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation
Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation R.Navaneethakrishnan Assistant Professor (SG) Bharathiyar College of Engineering and Technology, Karaikal, India.
Network Performance: Networks must be fast. What are the essential network performance metrics: bandwidth and latency
Network Performance: Networks must be fast What are the essential network performance metrics: bandwidth and latency Transmission media AS systems Input'signal'f(t) Has'bandwidth'B System'with'H(-) Output'signal'g(t)
Active Queue Management
Course of Multimedia Internet (Sub-course Reti Internet Multimediali ), AA 2010-2011 Prof. 6. Active queue management Pag. 1 Active Queue Management Active Queue Management (AQM) is a feature that can
AKAMAI WHITE PAPER. Delivering Dynamic Web Content in Cloud Computing Applications: HTTP resource download performance modelling
AKAMAI WHITE PAPER Delivering Dynamic Web Content in Cloud Computing Applications: HTTP resource download performance modelling Delivering Dynamic Web Content in Cloud Computing Applications 1 Overview
EVALUATING NETWORK BUFFER SIZE REQUIREMENTS
EVALUATING NETWORK BUFFER SIZE REQUIREMENTS for Very Large Data Transfers Michael Smitasin Lawrence Berkeley National Laboratory (LBNL) Brian Tierney Energy Sciences Network (ESnet) [ 2 ] Example Workflow
Network management and QoS provisioning - QoS in the Internet
QoS in the Internet Inernet approach is based on datagram service (best effort), so provide QoS was not a purpose for developers. Mainly problems are:. recognizing flows;. manage the issue that packets
A Compound TCP Approach for High-speed and Long Distance Networks
A Compound TCP Approach for High-speed and Long Distance Networks Kun Tan Jingmin Song Microsoft Research Asia Beijing, China {kuntan, jmsong, }@microsoft.com Qian Zhang DCS, Hong Kong Univ. of Sci & Tech
TCP over ATM Performance in NASA NREN and CTI
TCP over Performance in NASA NREN and CTI 1996 ARPA Workshop on Wide Area Performance Agatino Sciuto Computer Sciences Corporation email:[email protected] NASCOM 7/22/96 1 Background Introduction
Robust Router Congestion Control Using Acceptance and Departure Rate Measures
Robust Router Congestion Control Using Acceptance and Departure Rate Measures Ganesh Gopalakrishnan a, Sneha Kasera b, Catherine Loader c, and Xin Wang b a {[email protected]}, Microsoft Corporation,
TCP Westwood for Wireless
TCP Westwood for Wireless מבוא רקע טכני בקרת עומס ב- TCP TCP על קשר אלחוטי שיפור תפוקה עם פרוטוקול TCP Westwood סיכום.1.2.3.4.5 Seminar in Computer Networks and Distributed Systems Hadassah College Spring
Supporting VoIP in IEEE802.11 Distributed WLANs
Supporting VoIP in IEEE802.11 Distributed WLANs Zuo Liu Supervisor: Dr. Nick Filer July 2012 1 Voice VoIP Applications Constant Streaming Traffic Packetize interval usually 10-30 ms 8 160 bytes each packet
RFC 6349 Testing with TrueSpeed from JDSU Experience Your Network as Your Customers Do
RFC 6349 Testing with TrueSpeed from JDSU Experience Your Network as Your Customers Do RFC 6349 is the new transmission control protocol (TCP) throughput test methodology that JDSU co-authored along with
Network Performance Monitoring at Small Time Scales
Network Performance Monitoring at Small Time Scales Konstantina Papagiannaki, Rene Cruz, Christophe Diot Sprint ATL Burlingame, CA [email protected] Electrical and Computer Engineering Department University
Advanced Computer Networks. Introduction
Oriana Riva, Department of Computer Science ETH Zürich Advanced Computer Networks 263-3501-00 Introduction Patrick Stuedi, Qin Yin and Timothy Roscoe Spring Semester 2015 Information about the course http://www.systems.ethz.ch/courses/spring2014/acn
Berkeley Ninja Architecture
Berkeley Ninja Architecture ACID vs BASE 1.Strong Consistency 2. Availability not considered 3. Conservative 1. Weak consistency 2. Availability is a primary design element 3. Aggressive --> Traditional
Stability Criteria of RED with TCP Traffic
Stability Criteria of RED with TCP Traffic Technical Report Thomas Ziegler Serge Fdida Christof Brandauer {Thomas.Ziegler, Christof.Brandauer}@salzburgresearch.at [email protected] Université Pierre
# % # % & () # () + (, ( + + + ( (. /(. 0 + + ( (. /(. 0!12!3 &! 1. 4 ( /+ ) 0
! # % # % & () # () + (, ( + + + ( (. /(. 0 + + ( (. /(. 0!12!3 &! 1. 4 ( /+ ) 0 5 M21TCP: Overcoming TCP Incast Congestion in Data Centres Akintomide Adesanmi Lotfi Mhamdi School of Electronic & Electrical
Priority Queuing of Network Game Traffic over a DOCSIS Cable Modem Link
Priority Queuing of Network Game Traffic over a DOCSIS Cable Modem Link Jason But, Shaun Burriss and Grenville Armitage Centre for Advanced Internet Architectures Swinburne University of Technology Melbourne,
Active Queue Management (AQM) based Internet Congestion Control
Active Queue Management (AQM) based Internet Congestion Control October 1 2002 Seungwan Ryu ([email protected]) PhD Student of IE Department University at Buffalo Contents Internet Congestion Control
Networking part 3: the transport layer
Networking part 3: the transport layer Juliusz Chroboczek Université de Paris-Diderot (Paris 7) September 2011 Summary of the previous episodes Episode 1: switching, packet switching and the Internet.
LRU-RED: An active queue management scheme to contain high bandwidth flows at congested routers
LRU-RED: An active queue management scheme to contain high bandwidth flows at congested routers Smitha A. L. Narasimha Reddy Dept. of Elec. Engg., Texas A & M University, College Station, TX 77843-3128,
CSE 473 Introduction to Computer Networks. Exam 2 Solutions. Your name: 10/31/2013
CSE 473 Introduction to Computer Networks Jon Turner Exam Solutions Your name: 0/3/03. (0 points). Consider a circular DHT with 7 nodes numbered 0,,...,6, where the nodes cache key-values pairs for 60
Multipath TCP in Data Centres (work in progress)
Multipath TCP in Data Centres (work in progress) Costin Raiciu Joint work with Christopher Pluntke, Adam Greenhalgh, Sebastien Barre, Mark Handley, Damon Wischik Data Centre Trends Cloud services are driving
International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July-2015 1169 ISSN 2229-5518
International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July-2015 1169 Comparison of TCP I-Vegas with TCP Vegas in Wired-cum-Wireless Network Nitin Jain & Dr. Neelam Srivastava Abstract
Transport Layer Protocols
Transport Layer Protocols Version. Transport layer performs two main tasks for the application layer by using the network layer. It provides end to end communication between two applications, and implements
Experiences with Interactive Video Using TFRC
Experiences with Interactive Video Using TFRC Alvaro Saurin, Colin Perkins University of Glasgow, Department of Computing Science Ladan Gharai University of Southern California, Information Sciences Institute
Aggregate Traffic Performance with Active Queue Management and Drop from Tail
Aggregate Traffic Performance with Active Queue Management and Drop from Tail Gianluca Iannaccone Sprint ATL Adrian Court Burlingame CA 9400 [email protected] Martin May Activia Networks Parc of
Network traffic: Scaling
Network traffic: Scaling 1 Ways of representing a time series Timeseries Timeseries: information in time domain 2 Ways of representing a time series Timeseries FFT Timeseries: information in time domain
2 TCP-like Design. Answer
Homework 3 1 DNS Suppose you have a Host C, a local name server L, and authoritative name servers A root, A com, and A google.com, where the naming convention A x means that the name server knows about
Best-Case WiBro Performance for a Single Flow
Best-Case WiBro Performance for a Single Flow Shinae Woo, Keon Jang, Sangman Kim, Soohyun Cho, Jaehwa Lee, Youngseok Lee, and Sue Moon Computer Science Department, KAIST School of Computer Science & Engineering,
Linux TCP Implementation Issues in High-Speed Networks
Linux TCP Implementation Issues in High-Speed Networks D.J.Leith Hamilton Institute, Ireland www.hamilton.ie 1. Implementation Issues 1.1. SACK algorithm inefficient Packets in flight and not yet acknowledged
Applications. Network Application Performance Analysis. Laboratory. Objective. Overview
Laboratory 12 Applications Network Application Performance Analysis Objective The objective of this lab is to analyze the performance of an Internet application protocol and its relation to the underlying
TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) Internet Protocol (IP)
TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) *Slides adapted from a talk given by Nitin Vaidya. Wireless Computing and Network Systems Page
First Midterm for ECE374 03/24/11 Solution!!
1 First Midterm for ECE374 03/24/11 Solution!! Note: In all written assignments, please show as much of your work as you can. Even if you get a wrong answer, you can get partial credit if you show your
Gateway Strategies for VoIP Traffic over Wireless Multihop Networks
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL. 5, NO. 1, January 2011 24 Copyright c 2011 KSII Gateway Strategies for VoIP Traffic over Wireless Multihop Networks Kyungtae Kim 1, Dragoş Niculescu
Sockets vs. RDMA Interface over 10-Gigabit Networks: An In-depth Analysis of the Memory Traffic Bottleneck
Sockets vs. RDMA Interface over 1-Gigabit Networks: An In-depth Analysis of the Memory Traffic Bottleneck Pavan Balaji Hemal V. Shah D. K. Panda Network Based Computing Lab Computer Science and Engineering
Early Binding Updates and Credit-Based Authorization A Status Update
Status update New drafts Implementation Experimentation results Early Binding Updates and Credit-Based Authorization A Status Update Why Do We Need Enhancement? Mobile IPv6 Route Optimization uses return-routability
Buffer Management Schemes for Supporting TCP in Gigabit Routers with Per-flow Queueing
Buffer Management Schemes for Supporting TCP in Gigabit Routers with Per-flow Queueing B. Suter, T. V. Lakshman, D. Stiliadis, and A. K. Choudhury Abstract Recently, there has been much interest in using
Hybrid network traffic engineering system (HNTES)
Hybrid network traffic engineering system (HNTES) Zhenzhen Yan, Zhengyang Liu, Chris Tracy, Malathi Veeraraghavan University of Virginia and ESnet Jan 12-13, 2012 [email protected], [email protected] Project
TCP over Wireless Networks
TCP over Wireless Networks Raj Jain Professor of Computer Science and Engineering Washington University in Saint Louis Saint Louis, MO 63130 Audio/Video recordings of this lecture are available at: http://www.cse.wustl.edu/~jain/cse574-10/
TCP loss sensitivity analysis ADAM KRAJEWSKI, IT-CS-CE
TCP loss sensitivity analysis ADAM KRAJEWSKI, IT-CS-CE Original problem IT-DB was backuping some data to Wigner CC. High transfer rate was required in order to avoid service degradation. 4G out of... 10G
Friends, not Foes Synthesizing Existing Transport Strategies for Data Center Networks
Friends, not Foes Synthesizing Existing Transport Strategies for Data Center Networks Ali Munir Michigan State University Ghufran Baig, Syed M. Irteza, Ihsan A. Qazi, Alex X. Liu, Fahad R. Dogar Data Center
The Ecosystem of Computer Networks. Ripe 46 Amsterdam, The Netherlands
The Ecosystem of Computer Networks Ripe 46 Amsterdam, The Netherlands Silvia Veronese NetworkPhysics.com [email protected] September 2003 1 Agenda Today s IT challenges Introduction to Network
Distributed applications monitoring at system and network level
Distributed applications monitoring at system and network level Monarc Collaboration 1 Abstract Most of the distributed applications are presently based on architectural models that don t involve real-time
PACKET TRANSPORT MECHANISMS FOR DATA CENTER NETWORKS
PACKET TRANSPORT MECHANISMS FOR DATA CENTER NETWORKS A DISSERTATION SUBMITTED TO THE DEPARTMENT OF ELECTRICAL ENGINEERING AND THE COMMITTEE ON GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT