How$speedy$is$SPDY? Xiao%(Sophia)%Wang,$Aruna$Balasubramanian,$Arvind$ Krishnamurthy,$David$Wetherall$ University*of*Washington*

Size: px
Start display at page:

Download "How$speedy$is$SPDY? Xiao%(Sophia)%Wang,$Aruna$Balasubramanian,$Arvind$ Krishnamurthy,$David$Wetherall$ University*of*Washington*"

Transcription

1 HowspeedyisSPDY? Xiao%(Sophia)%Wang,ArunaBalasubramanian,Arvind Krishnamurthy,DavidWetherall University*of*Washington*

2 HTTP/1.1:(The(standard( to(load(web(pages HTTP/1.1'becomes(slow( for(rich,(modern(pages Google(developed(SPDY( to(make(the(web(faster A(StarCng(to(be(deployed( A(Basis(for(HTTP/2.0(now( being(standardized

3 HTTP/1.1:(The(standard( to(load(web(pages HowmuchbetterisSPDYthanHTTP? HTTP/1.1'becomes(slow( for(rich,(modern(pages Google(developed(SPDY( to(make(the(web(faster A(StarCng(to(be(deployed( A(Basis(for(HTTP/2.0(now( being(standardized

4 4/14/14

5 Client HTTP/1.1problems Server 4/14/14

6 Client HTTP/1.1problems Server OpenstoomanyTCPconnections 4/14/14

7 Client HTTP/1.1problems Server OpenstoomanyTCPconnections Initiatesobjecttransfersstrictlyby theclient 4/14/14

8 Client HTTP/1.1problems HTTP/1.1200OK\r\n.Date:Fri,21Mar2014\r\n.Server:Apache/2.2.26\r\n \r\n ******************** ******************** Uncompressed Compressed Server OpenstoomanyTCPconnections Initiatesobjecttransfersstrictlyby theclient CompressesonlyHTTPpayloads, notheaders 4/14/14

9 Client HTTP/1.1problems HTTP/1.1200OK\r\n.Date:Fri,21Mar2014\r\n.Server:Apache/2.2.26\r\n \r\n ******************** ******************** Uncompressed SPDYisproposedto Server Compressed addresstheseissues OpenstoomanyTCPconnections Initiatesobjecttransfersstrictlyby theclient CompressesonlyHTTPpayloads, notheaders 4/14/14

10 Client SPDY Server OpenstoomanyTCPconnections Multiplexesslicedframesintoa singletcpconnection 4/14/14

11 Client SPDY Server OpenstoomanyTCPconnections Multiplexesslicedframesintoa singletcpconnection PrioritizesWebobjects 4/14/14

12 Client SPDY Server Initiatesobjecttransfersstrictlyby theclient AllowsserverstoinitiateWeb objecttransfers 4/14/14

13 Client SPDY Compressed ******************** ******************** ******************** ******************** Compressed Server CompressesonlyHTTPpayloads, notheaders CompressesbothHTTPpayloads andheaders 4/14/14

14 HowwelldoesSPDYperform? SPDYhelps27%to60% SPDYsometimeshelpsand sometimeshurts. Overall,SPDYhelps<%10%.

15 HowwelldoesSPDYperform? SPDYhelps27%to60% SPDYsometimeshelpsand sometimeshurts. Overall,SPDYhelps<%10%. Measurementresultsconflict

16 Goals AsystematicstudyofSPDYthat Extensivelysweepstheparameterspace LinksSPDYperformancetounderlyingfactors Identifiesthedominantfactors SPDYv.s.HTTP/1.1

17 Challenge ManyfactorsexternaltoSPDYaffectSPDY Approach Isolatefactors,sweeptheparameterspace %Networkparameters %TCPsettings %Webpageeffects RTT Bandwidth Lossrate TCPinitialcongestionwindow Syntheticobjects Realobjects Realpages

18 Challenge Pageloadtimehashighvariance Approach Controlsourceofvariabilityby ` ` CDF HTTP SPDY Page load time (seconds) Variance:(0.5(second Difference:(0.02(second Experimentinginacontrolled%network% Usingour%emulator%insteadofbrowsers

19 Challenge Dependenciesbetweennetworkand browsercomputationaffectpageloads Objects No'browser Elapsedtime Objects Browser'computa<on Elapsedtime

20 Challenge Dependenciesbetweennetworkand browsercomputationaffectpageloads Objects No'browser Elapsedtime Objects Approach Preservedependencies.% Browser'computa<on Elapsedtime

21 Outline UnderstandingSPDY sperformancewith Syntheticobjects Realobjects Realpages

22 Outline UnderstandingSPDY sperformancewith Synthetic%objects% Realobjects Realpages

23 Extensivelysweepparameterspace Network parameters TCPsettings Synthetic objects Factors RTT Bandwidth Loss(rate TCP(IW Range 20ms,(100ms,(200ms 1Mbps,(10Mbps 0,(.5%,(1%,(2% 3,(10,(21,(32 Web(obj.(size 100B,(1K,(10K,(100K,(1M MakeHTTPrequests #(of(objects 2,(8,(16,(32,(64,(128,(512

24 LinkSPDYperformancetofactors!Decisiontreeanalysis small objsize large Sixfactors, Thousandsof datapoints loss loss low high low high SPDY #obj #obj #obj small large small large small SPDY SPDY large RTT RTT low high low RTT RTT high low high low high EQUAL EQUAL HTTP HTTP BW BW BW BW SPDY EQUAL SPDY EQUAL EQUAL HTTP IW IW EQUAL SPDY EQUAL SPDY

25 SPDYhelpsonsmallobjects obj%size Small%(<%1.5KB) loss low high SPDY #obj small large low RTT SPDY high low BW BW high low high SPDY EQUAL SPDY small IW large EQUAL SPDY UnlikeinHTTP,aTCPsegment cancarrymultiplewebobjects inspdy. % % % Explanation WhySPDY helps

26 SPDYhelpsonlargeobjects,lowloss low obj%size small RTT EQUAL BW low EQUAL small EQUAL Explanation Large%(>%1.5KB) Low%(<%0.5%) #obj large SPDY high high IW loss large SPDY InHTTP,Multipleconnections competewitheachother! Moreretransmissions% % %

27 SPDYhurtsonlargeobjects,highloss obj%size loss #obj RTT RTT BW HTTP EQUAL HTTP HTTP EQUAL Large%(>%1.5KB) High%(>=0.5%) small large low low high high low high Inasingleconnection,SPDY reducescwndmore aggressivelyunderloss. Explanation Time cwnd Time cwnd HTTP SPDY loss loss

28 SPDYhurtsonlargeobjects,highloss obj%size loss #obj RTT RTT BW HTTP EQUAL HTTP HTTP EQUAL Large%(>%1.5KB) High%(>=0.5%) small large low low high high low high Inasingleconnection,SPDY reducescwndmore aggressivelyunderloss. Explanation Time cwnd Time cwnd HTTP SPDY loss loss MostperformanceimpactofSPDYcomes fromasingletcpconnection.

29 Identifydominantfactors #obj BW RTT Importance? objsize loss IW

30 Identifydominantfactors objsize loss #obj more(important(than RTT BW IW PLTofSPDY/PLTofHTTP #obj PLTofSPDY/PLTofHTTP TCPIW RTT:200ms BW:10Mbps Loss:0 IW:3 objsize:10kb #obj:8 #objshowsatrend IWdoesn tshowatrend

31 DoesSPDYhelpstragglers? Loadtwoobjects (10Kand100K) Conn1 Conn2 BW cwnd HTTP/ Time BW cwnd 1 SPDY 1lessRTT Time Inourexperiments,wefindthatSPDYhelpslittle forstragglers.

32 DoesSPDYhelpstragglers? Conn1 Conn2 cwnd HTTP/ withcwnd Time SPDY Loadtwoobjects (10Kand100K) Thishypothesisisweaksinceitonlyargues BW BW cwnd 1 1lessRTT Time Inourexperiments,wefindthatSPDYhelpslittle forstragglers.

33 Outline UnderstandingSPDY sperformancewith Syntheticobjects Real%objects% Realpages

34 Syntheticobjects!Realobjects Network parameters TCPsettings Web objects Factors RTT Bandwidth Loss(rate TCP(IW Range 20ms,(100ms,(200ms 1Mbps,(10Mbps 0,(.5%,(1%,(2% 3,(10,(21,(32 Web(obj.(size 100B,(1K,(10K,(100K,(1M MakeHTTPrequests #(of(objects Top200Alexapages 2,(8,(16,(32,(64,(128,(512

35 1 CDF %% SPDYhelps HTTPhelps PLT of SPDY divided by PLT of HTTP SPDYhelps60%inthemediancase becauseitlargelyreducesretransmissions

36 Outline UnderstandingSPDY sperformancewith Syntheticobjects Realobjects Real%pages% Browser effects

37 Browsereffects html c.css f.js Elapsedtime html c.css f.js Elapsedtime Assumptionthatobjectsarefetchedat thesametimedoesnothold.

38 Eploadcapturesbrowsereffects Recorder:capturethedependencygraph Replayer:makenetworkrequestswhile simulatingthecomputationportions html c.css f.js Elapsedtime Realnetwork requests Wait Network% Computation%

39 Eploadcapturesbrowsereffects Recorder:capturethedependencygraph Replayer:makenetworkrequestswhile simulatingthecomputationportions html c.css f.js Elapsedtime Realnetwork requests Eploadmakesexperimentsreproducible Wait Network% Computation%

40 Realobjects!Realpages Network parameters TCPsettings Web objects Factors RTT Bandwidth Loss(rate TCP(IW Range 20ms,(100ms,(200ms 1Mbps,(10Mbps 0,(.5%,(1%,(2% 3,(10,(21,(32 Web(obj.(size 100B,(1K,(10K,(100K,(1M Top200Alexapages #(of(objects 2,(8,(16,(32,(64,(128,(512 EmulatepageloadswithEpload MakeHTTPrequests

41 1 CDF % SPDYhelps HTTPhelps PLT of SPDY divided by PLT of HTTP RTT=20ms Bandwidth=10Mbps SPDYhelpsmarginallybecause Computationanddependencies increasepltofbothspdyandhttp Throttledobjectfetchesresultinfewer retransmissionsinhttp

42 1 CDF % SPDYhelps HTTPhelps PLT of SPDY divided by PLT of HTTP RTT=20ms Bandwidth=10Mbps Dependenciesandcomputationinreal pageloadsreducetheimpactofspdy.

43 Otherexperimentsinthepaper UsingServerPush Experimentedwithnewpolicy Saves10%`30%%latencylikemod_spdybut pushes80%%lessdata WithSSL/TLS TestedSPDYandHTTPoverSSL/TLS Largerlatenciesbutsameconclusions

44 Conclusions WeexperimentedwithSPDYpageloadsover alargeparameterspace MostperformanceimpactofSPDYoverHTTP comesfromitssingletcpconnection Browsercomputationanddependenciesin realpagesreducetheimpactofspdy Toimprovefurther,weneedtorestructure thepageloadprocess

45

LISA: A Linked Slow-Start Algorithm for MPTCP draft-barik-mptcp-lisa-01

LISA: A Linked Slow-Start Algorithm for MPTCP draft-barik-mptcp-lisa-01 LISA: A Linked Slow-Start Algorithm for draft-barik-mptcp-lisa-1 Runa Barik (UiO), Simone Ferlin (SRL), Michael Welzl (UiO) Multipath TCP @96th IETF Meeting Berlin, Germany th July 16 IETF96 LISA: A Linked

More information

How speedy is SPDY? Xiao Sophia Wang, Aruna Balasubramanian, Arvind Krishnamurthy, and David Wetherall University of Washington Abstract

How speedy is SPDY? Xiao Sophia Wang, Aruna Balasubramanian, Arvind Krishnamurthy, and David Wetherall University of Washington Abstract How speedy is? Xiao Sophia Wang, Aruna Balasubramanian, Arvind Krishnamurthy, and David Wetherall University of Washington Abstract is increasingly being used as an enhancement to /.. To understand its

More information

Speeding up Web Page Loads with Shandian. Sophia Wang University of Washington

Speeding up Web Page Loads with Shandian. Sophia Wang University of Washington Speeding up Web Page Loads with Shandian Sophia Wang University of Washington Why is page load time (PLT) slow? 3/27/16 2

More information

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Araz Jangiaghdam Seminar Networks and Distributed Systems School of Engineering and Sciences Jacobs University Bremen Campus Ring 1,

More information

Why is My Website Running Slow? Edmond Chan & Feybian Yip Akamai Professional Services September 2018

Why is My Website Running Slow? Edmond Chan & Feybian Yip Akamai Professional Services September 2018 Why is My Website Running Slow? Edmond Chan & Feybian Yip Akamai Professional Services September 2018 Performance Vs. Conversion Desktop Tablet Mobile 4 Conversion rate (%) Conversion rate: 3 % visitors

More information

A Library and Proxy for SPDY

A Library and Proxy for SPDY A Library and Proxy for SPDY Interdisciplinary Project Andrey Uzunov Chair for Network Architectures and Services Department of Informatics Technische Universität München April 3, 2013 Andrey Uzunov (TUM)

More information

CSE/EE 461 HTTP and the Web

CSE/EE 461 HTTP and the Web CSE/EE 461 HTTP and the Web Last Time The Transport Layer Focus How does TCP share bandwidth? Topics AIMD Slow Start Application Presentation Session Transport Network Data Link Fast Retransmit / Fast

More information

ConfigTron: Tackling network diversity with heterogeneous configurations. Usama Naseer, Theophilus Benson Duke University

ConfigTron: Tackling network diversity with heterogeneous configurations. Usama Naseer, Theophilus Benson Duke University ConfigTron: Tackling network diversity with heterogeneous configurations Usama Naseer, Theophilus Benson Duke University 1 Diversity in network conditions Region Avg. RTT in ms Avg. packet loss % Trans-Atlantic

More information

PARCEL: Proxy Assisted BRowsing in Cellular networks for Energy and Latency reduction

PARCEL: Proxy Assisted BRowsing in Cellular networks for Energy and Latency reduction PARCEL: Proxy Assisted BRowsing in Cellular networks for Energy and Latency reduction Ashiwan Sivakumar 1, Shankaranarayanan PN 1, Vijay Gopalakrishnan 2, Seungjoon Lee 3*, Sanjay Rao 1 and Subhabrata

More information

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 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

More information

Demystifying Mobile Web Browsing under Multiple Protocols

Demystifying Mobile Web Browsing under Multiple Protocols 1 Demystifying Mobile Web Browsing under Multiple Protocols Yi Liu arxiv:1712.00237v1 [cs.ni] 1 Dec 2017 Abstract With the popularity of mobile devices, such as smartphones, tablets, users prefer visiting

More information

Investigating Forms of Simulating Web Traffic. Yixin Hua Eswin Anzueto Computer Science Department Worcester Polytechnic Institute Worcester, MA

Investigating Forms of Simulating Web Traffic. Yixin Hua Eswin Anzueto Computer Science Department Worcester Polytechnic Institute Worcester, MA Investigating Forms of Simulating Web Traffic Yixin Hua Eswin Anzueto Computer Science Department Worcester Polytechnic Institute Worcester, MA Outline Introduction Web Traffic Characteristics Web Traffic

More information

FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web

FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web What s the impact of web latency? 100ms 1% Delay sales Source : https://speakerdeck.com/deanohume/faster-mobilewebsites! 100ms 1% Delay revenue

More information

CS132/EECS148 - Instructor: Karim El Defrawy Midterm Spring 2013 Time: 1hour May 2nd, 2013

CS132/EECS148 - Instructor: Karim El Defrawy Midterm Spring 2013 Time: 1hour May 2nd, 2013 CS132/EECS148 - Instructor: Karim El Defrawy Midterm Spring 2013 : 1hour May 2nd, 2013 Total Points: 25 Attempt all problems. Problem #1: (5 points, ½ point each) Choose only one answer. You will not receive

More information

Fig. 2: Time gap between retransmission clusters for different connections.

Fig. 2: Time gap between retransmission clusters for different connections. Domain-Sharding for Faster HTTP/2 in Lossy Cellular Networks Utkarsh Goel, Moritz Steiner, Mike P. Wittie, Stephen Ludin, Martin Flack Akamai Technologies, Inc. {ugoel, moritz, mflack, sludin}@akamai.com

More information

CPSC 441 COMPUTER COMMUNICATIONS MIDTERM EXAM SOLUTION

CPSC 441 COMPUTER COMMUNICATIONS MIDTERM EXAM SOLUTION CPSC 441 COMPUTER COMMUNICATIONS MIDTERM EXAM SOLUTION Department of Computer Science University of Calgary Professor: Carey Williamson March 2, 2012 This is a CLOSED BOOK exam. Textbooks, notes, laptops,

More information

Assignment front sheet

Assignment front sheet Criteria reference To achieve the criteria the evidence must show that the student is able to: Task no. Page numbers P1 Outline the web architecture and components which enable internet and web functionality.

More information

Application Layer: The Web and HTTP Sec 2.2 Prof Lina Battestilli Fall 2017

Application Layer: The Web and HTTP Sec 2.2 Prof Lina Battestilli Fall 2017 CSC 401 Data and Computer Communications Networks Application Layer: The Web and HTTP Sec 2.2 Prof Lina Battestilli Fall 2017 Outline Application Layer (ch 2) 2.1 principles of network applications 2.2

More information

L2: Bandwidth and Latency. Hui Chen, Ph.D. Dept. of Engineering & Computer Science Virginia State University Petersburg, VA 23806

L2: Bandwidth and Latency. Hui Chen, Ph.D. Dept. of Engineering & Computer Science Virginia State University Petersburg, VA 23806 L2: Bandwidth and Latency Hui Chen, Ph.D. Dept. of Engineering & Computer Science Virginia State University Petersburg, VA 23806 1 Acknowledgements Some pictures used in this presentation were obtained

More information

Lec 17 Multimedia Transport: RTP, TCP/HTTP and QUIC

Lec 17 Multimedia Transport: RTP, TCP/HTTP and QUIC Multimedia Communication Lec 17 Multimedia Transport: RTP, TCP/HTTP and QUIC Zhu Li Course Web: http://l.web.umkc.edu/lizhu/teaching/2016sp.video-communication/main.html Z. Li, Multimedia Communciation,

More information

Modeling HTTP/2 Speed from HTTP/1 Traces

Modeling HTTP/2 Speed from HTTP/1 Traces Modeling HTTP/2 Speed from HTTP/1 Traces Kyriakos Zarifis 1, Mark Holland 2, Manish Jain 2, Ethan Katz-Bassett 1, Ramesh Govindan 1 1 University of Southern California 2 Akamai Technologies Abstract. With

More information

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 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

More information

2- Application Level Protocols HTTP 1.0/1.1/2

2- Application Level Protocols HTTP 1.0/1.1/2 2- Application Level Protocols HTTP 1.0/1.1/2 HTTP, (HyperText Transfer Protocol) Basis for fetching Web pages request Network CSE 461 University of Washington 2 Sir Tim Berners-Lee (1955 ) Inventor of

More information

BBR Congestion Control: IETF 100 Update: BBR in shallow buffers

BBR Congestion Control: IETF 100 Update: BBR in shallow buffers BBR Congestion Control: IETF 100 Update: BBR in shallow buffers Neal Cardwell, Yuchung Cheng, C. Stephen Gunn, Soheil Hassas Yeganeh Ian Swett, Jana Iyengar, Victor Vasiliev Van Jacobson https://groups.google.com/d/forum/bbr-dev

More information

Investigating the Use of Synchronized Clocks in TCP Congestion 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

More information

SPDY - A Web Protocol. Mike Belshe Velocity, Dec 2009

SPDY - A Web Protocol. Mike Belshe Velocity, Dec 2009 SPDY - A Web Protocol Mike Belshe Velocity, Dec 2009 What is SPDY? Concept SPDY is an application layer protocol for transporting content over the web with reduced latency. Basic Features 1. Multiplexed

More information

Over Lesson 2 6 Identify the type of function represented by the graph. Identify the type of function represented by the graph. Over Lesson 2 6 Identi

Over Lesson 2 6 Identify the type of function represented by the graph. Identify the type of function represented by the graph. Over Lesson 2 6 Identi Five-Minute Check (over Lesson 2 6) CCSS Then/Now New Vocabulary Key Concept: Parent Functions Example 1: Identify a Function Given the Graph Example 2: Describe and Graph Translations Example 3: Describe

More information

EXAM TCP/IP NETWORKING Duration: 3 hours

EXAM TCP/IP NETWORKING Duration: 3 hours SCIPER: First name: Family name: EXAM TCP/IP NETWORKING Duration: 3 hours Jean-Yves Le Boudec January 2017 INSTRUCTIONS 1. Write your solution into this document and return it to us (you do not need to

More information

Transmission Control Protocol. ITS 413 Internet Technologies and Applications

Transmission Control Protocol. ITS 413 Internet Technologies and Applications Transmission Control Protocol ITS 413 Internet Technologies and Applications Contents Overview of TCP (Review) TCP and Congestion Control The Causes of Congestion Approaches to Congestion Control TCP Congestion

More information

Applica8on Layer Overview

Applica8on Layer Overview Introduc8on to Computer Networks Applica8on Layer Overview Computer Science & Engineering Where we are in the Course Star8ng the Applica8on Layer! Builds distributed network services (DNS, Web) on Transport

More information

Implementation Experiments on HighSpeed and Parallel TCP

Implementation Experiments on HighSpeed and Parallel TCP Implementation Experiments on HighSpeed and TCP Zongsheng Zhang Go Hasegawa Masayuki Murata Osaka University Outline Introduction Background of and g Why to evaluate in a test-bed network A refined algorithm

More information

One More Bit Is Enough

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

More information

Congestion Control For Coded TCP. Doug Leith

Congestion Control For Coded TCP. Doug Leith Congestion Control For Coded TCP Doug Leith Outline Why do error-correction coding at the transport layer? Congestion control on lossy paths Implementation & performance measurements Why error-correction

More information

Experimental Evaluation of Transport Services CoAP, HTTP and SPDY for Internet of Things

Experimental Evaluation of Transport Services CoAP, HTTP and SPDY for Internet of Things Experimental Evaluation of Transport Services CoAP, HTTP and SPDY for Internet of Things Laila Daniel, Markku Kojo and Mikael Latvala Department of Computer Science University of Helsinki Mosa Consulting,

More information

CS 268: Lecture 7 (Beyond TCP Congestion Control)

CS 268: Lecture 7 (Beyond TCP Congestion Control) Outline CS 68: Lecture 7 (Beyond TCP Congestion Control) TCP-Friendly Rate Control (TFRC) explicit Control Protocol Ion Stoica Computer Science Division Department of Electrical Engineering and Computer

More information

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?

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.

More information

EXAM TCP/IP NETWORKING Duration: 3 hours With Solutions

EXAM TCP/IP NETWORKING Duration: 3 hours With Solutions SCIPER: First name: Family name: EXAM TCP/IP NETWORKING Duration: 3 hours With Solutions Jean-Yves Le Boudec January 2013 INSTRUCTIONS 1. Write your solution into this document and return it to us (you

More information

SPDY. HTML5. Make the web fast

SPDY. HTML5. Make the web fast SPDY. HTML5 Make the web fast Christian Horny GOOGLE: @igrigorik Devcon Usability Engineering DELAY USER REACTION 0-100 ms Instant 100 300 ms Feels sluggish 300-1000 ms Machine is working... 1 s+ Mental

More information

Internet Content Distribution

Internet Content Distribution Internet Content Distribution Chapter 1: Introduction Jussi Kangasharju Chapter Outline Introduction into content distribution Basic concepts TCP DNS HTTP Outline of the rest of the course Kangasharju:

More information

Flash: an efficient and portable web server

Flash: an efficient and portable web server Flash: an efficient and portable web server High Level Ideas Server performance has several dimensions Lots of different choices on how to express and effect concurrency in a program Paper argues that

More information

Web Quality of Experience

Web Quality of Experience A Platform for Crowdsourcing Web Quality of Experience Measurements Matteo Varvello Telefónica Research Jeremy Blackburn Telefónica Research David Naylor Carnegie Mellon University Dina Papagiannaki Google

More information

TCP WISE: One Initial Congestion Window Is Not Enough

TCP WISE: One Initial Congestion Window Is Not Enough TCP WISE: One Initial Congestion Window Is Not Enough Xiaohui Nie $, Youjian Zhao $, Guo Chen, Kaixin Sui, Yazheng Chen $, Dan Pei $, MiaoZhang, Jiyang Zhang $ 1 Motivation Web latency matters! latency

More information

Advanced Computer Networking. CYBR 230 Jeff Shafer University of the Pacific QUIC

Advanced Computer Networking. CYBR 230 Jeff Shafer University of the Pacific QUIC CYBR 230 Jeff Shafer University of the Pacific QUIC 2 It s a Google thing. (Originally) 3 Google Engineering Motivations Goal: Decrease end-user latency on web To increase user engagement So they see more

More information

Master s Thesis. TCP Congestion Control Mechanisms for Achieving Predictable Throughput

Master s Thesis. TCP Congestion Control Mechanisms for Achieving Predictable Throughput Master s Thesis Title TCP Congestion Control Mechanisms for Achieving Predictable Throughput Supervisor Prof. Hirotaka Nakano Author Kana Yamanegi February 14th, 2007 Department of Information Networking

More information

Detects Potential Problems. Customizable Data Columns. Support for International Characters

Detects Potential Problems. Customizable Data Columns. Support for International Characters Home Buy Download Support Company Blog Features Home Features HttpWatch Home Overview Features Compare Editions New in Version 9.x Awards and Reviews Download Pricing Our Customers Who is using it? What

More information

HTTP/2: Ease the Transition, Remove Implementation Barriers. Robert Haynes Ronnie Dockery

HTTP/2: Ease the Transition, Remove Implementation Barriers. Robert Haynes Ronnie Dockery HTTP/2: Ease the Transition, Remove Implementation Barriers Robert Haynes Ronnie Dockery No One Likes Slow 3 No One Likes Slow 74% of users will leave a slow website after just 5 seconds or less 3 No One

More information

Computer Network Fundamentals Spring Week 10 Congestion Control Andreas Terzis

Computer Network Fundamentals Spring Week 10 Congestion Control Andreas Terzis Computer Network Fundamentals Spring 2008 Week 10 Congestion Control Andreas Terzis Outline Congestion Control TCP Congestion Control CS 344/Spring08 2 What We Know We know: How to process packets in a

More information

Congestion Avoidance and Control. Rohan Tabish and Zane Ma

Congestion Avoidance and Control. Rohan Tabish and Zane Ma Congestion Avoidance and Control Rohan Tabish and Zane Ma TCP is self-clocking Self-clocking systems should be robust Congestion collapse Internet had first of what became a series of congestion collapses

More information

IBX esourcing Client Side Requirements

IBX esourcing Client Side Requirements IBX esourcing 8.0.2 Client Side Requirements Capgemini Capgemini Procurement Services Table of contents 1 Introduction 2 Minimum System Requirements for End Users 3 Recommended Default Browser Settings

More information

Q U E S T I O N 3 In the current version of IP (IPv4), the use of TCP and UDP headers differ in which of the following ways?

Q U E S T I O N 3 In the current version of IP (IPv4), the use of TCP and UDP headers differ in which of the following ways? Preview Test: HW 2 Test Information Description Chapter 2 Instructions Due:Oct. 17 5 PM Multiple Attempts Not allowed. This test can only be taken once. Force Completion This test can be saved and resumed

More information

Investigating the Use of Synchronized Clocks in TCP Congestion Control

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

More information

CS268: Beyond TCP Congestion Control

CS268: Beyond TCP Congestion Control TCP Problems CS68: Beyond TCP Congestion Control Ion Stoica February 9, 004 When TCP congestion control was originally designed in 1988: - Key applications: FTP, E-mail - Maximum link bandwidth: 10Mb/s

More information

Client/Server Computing

Client/Server Computing CS 312 Internet Concepts Client/Server Computing Dr. Michele Weigle Department of Computer Science Old Dominion University mweigle@cs.odu.edu http://www.cs.odu.edu/~mweigle/cs312-f11 1 Applications and

More information

TCP Congestion Control 65KB W

TCP Congestion Control 65KB W TCP Congestion Control 65KB W TO 3DA 3DA TO 0.5 0.5 0.5 0.5 3 3 1 SS SS CA SS CA TCP s Congestion Window Maintenance TCP maintains a congestion window (cwnd), based on packets Sender s window is limited

More information

Review of Previous Lecture

Review of Previous Lecture Review of Previous Lecture Network access and physical media Internet structure and ISPs Delay & loss in packet-switched networks Protocol layers, service models Some slides are in courtesy of J. Kurose

More information

Tuning RED for Web Traffic

Tuning RED for Web Traffic Tuning RED for Web Traffic Mikkel Christiansen, Kevin Jeffay, David Ott, Donelson Smith UNC, Chapel Hill SIGCOMM 2000, Stockholm subsequently IEEE/ACM Transactions on Networking Vol. 9, No. 3 (June 2001)

More information

Computer Networking

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

More information

Exam - Final. CSCI 1680 Computer Networks Fonseca. Closed Book. Maximum points: 100 NAME: 1. TCP Congestion Control [15 pts]

Exam - Final. CSCI 1680 Computer Networks Fonseca. Closed Book. Maximum points: 100 NAME: 1. TCP Congestion Control [15 pts] CSCI 1680 Computer Networks Fonseca Exam - Final Due: 11:00am, May 10th, 2012 Closed Book. Maximum points: 100 NAME: 1. TCP Congestion Control [15 pts] a. TCP Tahoe and Reno have two congestion-window

More information

Experimental Experiences with Emulating Multihop Path using Dummynet

Experimental Experiences with Emulating Multihop Path using Dummynet Experimental Experiences with Emulating Multihop Path using Dummynet Thuy T.T. Nguyen Introduction Motivations How to emulate multihop path using Dummynet Test setup & Results Analysis Conclusions 13/04/2005

More information

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK [r.tasker@dl.ac.uk] DataTAG is a project sponsored by the European Commission - EU Grant IST-2001-32459

More information

Towards Understanding TCP Performance on LTE/EPC Mobile Networks!

Towards Understanding TCP Performance on LTE/EPC Mobile Networks! Towards Understanding TCP Performance on LTE/EPC Mobile Networks! Binh Nguyen1, Arijit Banerjee1, Vijay Gopalakrishnan2, Sneha Kasera1, Seungjoon Lee2, Aman Shaikh2, and Jacobus (Kobus) Van der Merwe1

More information

Manage Files. Accessing Manage Files

Manage Files. Accessing Manage Files 1 Manage Files The Manage Files tool is a file management system for your course. You can use this tool to organize and upload files associated with your course offering. We recommend that you organize

More information

Improving HTTP Latency

Improving HTTP Latency Improving HTTP Latency Venkata N. Padmanabhan (U.C.Berkeley) Jeffrey C. Mogul (DEC Western Research Lab.) The Second International WWW Conference, Chicago October 17-20, 1994 [1] Outline Motivation Sources

More information

20486-Developing ASP.NET MVC 4 Web Applications

20486-Developing ASP.NET MVC 4 Web Applications Course Outline 20486-Developing ASP.NET MVC 4 Web Applications Duration: 5 days (30 hours) Target Audience: This course is intended for professional web developers who use Microsoft Visual Studio in an

More information

A New Internet? RIPE76 - Marseille May Jordi Palet

A New Internet? RIPE76 - Marseille May Jordi Palet A New Internet? RIPE76 - Marseille May 2018 Jordi Palet (jordi.palet@theipv6company.com) -1 (a quick) Introduction to HTTP/2, QUIC and DOH and more RIPE76 - Marseille May 2018 Jordi Palet (jordi.palet@theipv6company.com)

More information

BBR Congestion Control Work at Google IETF 101 Update

BBR Congestion Control Work at Google IETF 101 Update BBR Congestion Control Work at Google IETF 101 Update Neal Cardwell, Yuchung Cheng, C. Stephen Gunn, Soheil Hassas Yeganeh Ian Swett, Jana Iyengar, Victor Vasiliev Priyaranjan Jha, Yousuk Seung Van Jacobson

More information

Tech Note. ConnectWise PSA Integration

Tech Note. ConnectWise PSA Integration GMS ConnectWise PSA Integration Contents Contents... 1 Overview... 1 Prerequisites... 1 GMS Configuration... 1 ConnectWise Configuration... 4 ConnectWise Customer Configurations... 7 Verification of Reporting

More information

An Argument for Increasing TCP s Initial Congestion Window

An Argument for Increasing TCP s Initial Congestion Window An Argument for Increasing TCP s Initial Congestion Window Nandita Dukkipati, Tiziana Refice, Yuchung Cheng, Jerry Chu Tom Herbert, Amit Agarwal, Arvind Jain and Natalia Sutin Google Inc. Mountain View,

More information

Congestion Control Without a Startup Phase

Congestion Control Without a Startup Phase Congestion Control Without a Startup Phase Dan Liu 1, Mark Allman 2, Shudong Jin 1, Limin Wang 3 1. Case Western Reserve University, 2. International Computer Science Institute, 3. Bell Labs PFLDnet 2007

More information

TCP so far Computer Networking Outline. How Was TCP Able to Evolve

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

More information

Riptide: Jump Starting Back-Office Connections in Cloud Systems

Riptide: Jump Starting Back-Office Connections in Cloud Systems Riptide: Jump Starting Back-Office Connections in Cloud Systems Marcel Flores - Northwestern University Amir R. Khakpour - Verizon Digital Media Services Harkeerat Bedi - Verizon Digital Media Services

More information

CMSC 417. Computer Networks Prof. Ashok K Agrawala Ashok Agrawala. October 30, 2018

CMSC 417. Computer Networks Prof. Ashok K Agrawala Ashok Agrawala. October 30, 2018 CMSC 417 Computer Networks Prof. Ashok K Agrawala 2018 Ashok Agrawala October 30, 2018 Message, Segment, Packet, and Frame host host HTTP HTTP message HTTP TCP TCP segment TCP router router IP IP packet

More information

Model-based Measurements Of Network Loss

Model-based Measurements Of Network Loss Model-based Measurements Of Network Loss June 28, 2013 Mirja Kühlewind mirja.kuehlewind@ikr.uni-stuttgart.de Universität Stuttgart Institute of Communication Networks and Computer Engineering (IKR) Prof.

More information

IETF96 INFO SECURE CONTENT DELEGATION (BLIND CACHE)

IETF96 INFO SECURE CONTENT DELEGATION (BLIND CACHE) IETF96 INFO SECURE CONTENT DELEGATION (BLIND CACHE) TODAY S AGENDA WHAT IT IS WHERE WE ARE (SPECS&IMPL) DISCUSSION DRAFT-THOMSON-HTTP-SCD SECURE CONTENT DELEGATION SPLIT CONTENT AND METADATA AND HOST CONTENT

More information

BEx Front end Performance

BEx Front end Performance BUSINESS INFORMATION WAREHOUSE BEx Front end Performance Performance Analyses of BEx Analyzer and Web Application in the Local and Wide Area Networks Environment Document Version 1.1 March 2002 Page 2

More information

Measuring end-to-end bandwidth with Iperf using Web100

Measuring end-to-end bandwidth with Iperf using Web100 Measuring end-to-end bandwidth with Iperf using Web Ajay Tirumala Les Cottrell Tom Dunigan CS Dept. SLAC CSMD University of Illinois Stanford University ORNL Urbana, IL 6181 Menlo Park, CA 9425 Oak Ridge,

More information

Lab 2. All datagrams related to favicon.ico had been ignored. Diagram 1. Diagram 2

Lab 2. All datagrams related to favicon.ico had been ignored. Diagram 1. Diagram 2 Lab 2 All datagrams related to favicon.ico had been ignored. Diagram 1 Diagram 2 1. Is your browser running HTTP version 1.0 or 1.1? What version of HTTP is the server running? According to the diagram

More information

Implementing Citrix XenApp 5.0 for Windows Server 2008

Implementing Citrix XenApp 5.0 for Windows Server 2008 Citrix 1Y0-A05 Implementing Citrix XenApp 5.0 for Windows Server 2008 Version: 5.0 Topic 1, Volume A QUESTION NO: 1 An administrator currently has Secure Gateway and web interface on the same server in

More information

Improving the Robustness of TCP to Non-Congestion Events

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

More information

Order Central Requirements 08/04/2009

Order Central Requirements 08/04/2009 Order Central Requirements 08/04/2009 Contents: Contents:... 1 Table of Figures:... 1 Order Central Architecture... 2 Database:... 2 :... 3 Server:... 3 Browsers:... 3 Minimum Recommended Setup:... 4 Optimum

More information

A transport-layer approach for achieving predictable throughput for Internet applications

A transport-layer approach for achieving predictable throughput for Internet applications Seventh International Conference on Networking A transport-layer approach for achieving predictable throughput for Internet applications Go Hasegawa, Kana Yamanegi and Masayuki Murata Graduate School of

More information

HTTP/2: What You Need to Know. Robert

HTTP/2: What You Need to Know. Robert HTTP/2: What You Need to Know Robert Boedigheimer @boedie About Me Web developer since 1995 Pluralsight Author 3 rd Degree Black Belt, Tae Kwon Do ASP.NET MVP boedie@outlook.com @boedie weblogs.asp.net/boedie

More information

Firmware Update Procedure by HTTP Server (FIRMWARE1/FIRMWARE2/FIRMWARE3/FIRMWARE4) Rev.5.0

Firmware Update Procedure by HTTP Server (FIRMWARE1/FIRMWARE2/FIRMWARE3/FIRMWARE4) Rev.5.0 Firmware Update Procedure by HTTP Server (FIRMWARE1/FIRMWARE2/FIRMWARE3/FIRMWARE4) Rev.5.0 Read This Manual Before Attempting Firmware Update This manual describes the procedure for updating firmware of

More information

USER GUIDE PowerPhoto CRM

USER GUIDE PowerPhoto CRM USER GUIDE PowerPhoto CRM 2013 2015 Contents CONFIGURING POWERPHOTO Add, Remove, Set Default, Set Record Image, Thumbnail & View Text Display Original File Save Original File Resize Photo Resize Height

More information

Chapter 3 Transport Layer

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

More information

a. (4pts) What general information is contained in a LSR-PDU update that A might send?

a. (4pts) What general information is contained in a LSR-PDU update that A might send? B1: Networks (25 points) Link State Routing (LSR). (Hint: flooding and Dijkstra s Algorithm). Assume Router A has physical links to Routers W, X, Y, Z. a. (4pts) What general information is contained in

More information

Content Delivery on the Web: HTTP and CDNs

Content Delivery on the Web: HTTP and CDNs Content Delivery on the Web: HTTP and CDNs Mark Handley UCL Computer Science CS 3035/GZ01 Outline The Web: HTTP and caching The Hypertext Transport Protocol: HTTP HTTP performance Persistent and concurrent

More information

Connecting the World:

Connecting the World: Connecting the World: A look inside Facebook s Networking Infrastructure Arun Moorthy arunm@fb.com https://fb.me/arun.moorthy about:me B. Tech CSE - Indian Institute of Technology, 1997 MSCS, University

More information

On HTTP Performance in IoT Applications: An Analysis of Latency and Throughput

On HTTP Performance in IoT Applications: An Analysis of Latency and Throughput On HTTP Performance in IoT Applications: An Analysis of Latency and Throughput Wolfgang Bziuk, Cao Vien Phung, Jasenka Dizdarević and Admela Jukan Technische Universität Braunschweig, Germany Email:{w.bziuk,

More information

IMPORT GUIDE. PowerPhoto CRM 2013/2015

IMPORT GUIDE. PowerPhoto CRM 2013/2015 IMPORT GUIDE PowerPhoto CRM 2013/2015 Contents Import Register Configure Add, Remove, Set Default, Set Record Image, Thumbnail & View Text Display Original File Save Original File Resize Photo Resize Height

More information

Congestion Control for High Bandwidth-Delay Product Networks

Congestion Control for High Bandwidth-Delay Product Networks Congestion Control for High Bandwidth-Delay Product Networks Presented by: Emad Shihab Overview Introduce the problem of XCP (what the protocol achieves) (how the protocol achieves this) The problem! TCP

More information

Measuring end-to-end bandwidth with Iperf using Web100

Measuring end-to-end bandwidth with Iperf using Web100 Presented at Passive and Active Monitoring Workshop (PAM 23), 4/6/23-4/8/23, San Diego, CA, USA SLAC-PUB-9733 April 23 Measuring end-to-end bandwidth with Iperf using Web1 Ajay Tirumala Les Cottrell Tom

More information

Module 6 Node.js and Socket.IO

Module 6 Node.js and Socket.IO Module 6 Node.js and Socket.IO Module 6 Contains 2 components Individual Assignment and Group Assignment Both are due on Wednesday November 15 th Read the WIKI before starting Portions of today s slides

More information

TCP Westwood: Efficient Transport for High-speed wired/wireless Networks

TCP Westwood: Efficient Transport for High-speed wired/wireless Networks TCP Westwood: Efficient Transport for High-speed wired/wireless Networks Mario Gerla, Medy Sanadidi, Ren Wang and Massimo Valla UCLA Computer Science 1 Outline 1. TCP Overview 2. Bandwidth Estimation and

More information

Module 10 Frame Relay and ATM

Module 10 Frame Relay and ATM Module 10 Frame Relay and ATM Lesson 35 ATM: Virtual Path, Virtual Channel. ATM Adaptation Layer (AAL) 10.3.1 VIRTUAL PATH AND VIRTUAL CHANNEL Connection between two endpoints is accomplished through virtual

More information

Understanding Multistreaming for Web Traffic: An Experimental Study

Understanding Multistreaming for Web Traffic: An Experimental Study Understanding Multistreaming for Web Traffic: An Experimental Study M. Rajiullah, A. C. Mohideen, F. Weinrank, R. Secchi, G. Fairhurst and A. Brunstrom Karlstad University, Karlstad, Sweden {mohammad.rajiullah,

More information

VISION. Network Assessment Tool User Guide

VISION. Network Assessment Tool User Guide VISION Network Assessment Tool User Guide Vision Network Assessment Tool - Presentation Ensuring our customers a reliable service is our first priority. ArkadinVision is designed to guarantee they will

More information

Overview Content Delivery Computer Networking Lecture 15: The Web Peter Steenkiste. Fall 2016

Overview Content Delivery Computer Networking Lecture 15: The Web Peter Steenkiste. Fall 2016 Overview Content Delivery 15-441 15-441 Computer Networking 15-641 Lecture 15: The Web Peter Steenkiste Fall 2016 www.cs.cmu.edu/~prs/15-441-f16 Web Protocol interactions HTTP versions Caching Cookies

More information

Customized Firmware: R0: V5.4.5 Build R1: V5.4.0 Build R0 PTZ: V5.3.0 Build

Customized Firmware: R0: V5.4.5 Build R1: V5.4.0 Build R0 PTZ: V5.3.0 Build Product: R0, R1 Platform IPC camera, R0 Platform Speed Dome Page: 1 of 9 Supported List R0 IPC: DS-2CD2012-I, DS-2CD2022-I, DS-2CD2032-I, DS-2CD2012F-I(W), DS-2CD2022F-I(W), DS-2CD2032F-I(W), DS- 2CD2112F-I(S)(W),

More information

Congestion Control for High-Bandwidth-Delay-Product Networks: XCP vs. HighSpeed TCP and QuickStart

Congestion Control for High-Bandwidth-Delay-Product Networks: XCP vs. HighSpeed TCP and QuickStart Congestion Control for High-Bandwidth-Delay-Product Networks: XCP vs. HighSpeed TCP and QuickStart Sally Floyd September 11, 2002 ICIR Wednesday Lunch 1 Outline: Description of the problem. Description

More information