Models. Motivation Timing Diagrams Metrics Evaluation Techniques. TOC Models

Similar documents
Models. One Packet. Timing. Illustration. Examples UCB. Models EECS 122. P bits. Motivation Timing Diagrams Metrics Evaluation Techniques

Performance Modeling

Outline. EE 122: Networks Performance & Modeling. Outline. Motivations. Definitions. Timing Diagrams. Ion Stoica TAs: Junda Liu, DK Moon, David Zats

Midterm Review EECS 122. University of California Berkeley

Midterm Review EECS 122. University of California Berkeley

Midterm Review. Topics. Review: Network. Review: Network. Layers & Protocols. Review: Network WAN. Midterm Review EECS 122 EECS 122

Transport Layer PREPARED BY AHMED ABDEL-RAOUF

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?

CPSC 441 Tutorial-1. Department of Computer Science University of Calgary

Network Management & Monitoring

Computer Networks. Homework #4: No Grading

ECE 610: Homework 4 Problems are taken from Kurose and Ross.

ETSF10 Internet Protocols Transport Layer Protocols

r=0.1 r=0.5 r=1 Figure 2.3: Throughput of ABP vs. PER

estadium Project Lab 2: Iperf Command

Hybrid Control and Switched Systems. Lecture #17 Hybrid Systems Modeling of Communication Networks

End-to-End Mechanisms for QoS Support in Wireless Networks

Network Management & Monitoring Network Delay

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

2.993: Principles of Internet Computing Quiz 1. Network

Topics. TCP sliding window protocol TCP PUSH flag TCP slow start Bulk data throughput

CPE 548 Exam #1 (50 pts) February 17, 2016

ETSN01 Exam Solutions

CS 344/444 Computer Network Fundamentals Final Exam Solutions Spring 2007

Lecture 9 Congestion Control: Part II. EECS 122 University of California Berkeley

EE122 MIDTERM EXAM: Scott Shenker, Ion Stoica

Why Your Application only Uses 10Mbps Even the Link is 1Gbps?

cs/ee 143 Fall

Computer Networks. Homework #3: Due 11 November 2015 (Wednesday) Just before Midterm#2!

CS/ECE 438: Communication Networks for Computers Spring 2018 Midterm Examination Online

ECEN Final Exam Fall Instructor: Srinivas Shakkottai

Appendix B. Standards-Track TCP Evaluation

c) With the selective repeat protocol, it is possible for the sender to receive an ACK for a packet that falls outside of its current window.

Chapter 3 outline. 3.5 Connection-oriented transport: TCP. 3.6 Principles of congestion control 3.7 TCP congestion control

Teletraffic theory (for beginners)

PRACTICE QUESTIONS ON RESOURCE ALLOCATION

II. Principles of Computer Communications Network and Transport Layer

Congestion control in TCP

Cover sheet for Assignment 3

CS Transport. Outline. Window Flow Control. Window Flow Control

Homework 1. Question 1 - Layering. CSCI 1680 Computer Networks Fonseca

Communication Networks

A Hybrid Systems Modeling Framework for Fast and Accurate Simulation of Data Communication Networks. Motivation

On Network Dimensioning Approach for the Internet

An Efficient Scheduling Scheme for High Speed IEEE WLANs

Chapter I: Introduction

Congestion Control in TCP

8. TCP Congestion Control

Investigating the Use of Synchronized Clocks in TCP Congestion Control

EPON Scheduling Protocol Requirements

Computer Networking. Introduction. Quintin jean-noël Grenoble university

Reliable Transport I: Concepts and TCP Protocol

First Exam for ECE671 Spring /22/18

Lecture 6. TCP services. Internet Transport Layer: introduction to the Transport Control Protocol (TCP) A MUCH more complex transport

Handles all kinds of traffic on a single network with one class

Congestion Avoidance

CSE 473 Introduction to Computer Networks. Final Exam. Your name here: 12/17/2012

Communication and Networks. Problems

Latency on a Switched Ethernet Network

Problem-1: Solution-1:

CS 421: COMPUTER NETWORKS SPRING FINAL May 21, minutes

CS244a: An Introduction to Computer Networks

Reliable Transport I: Concepts and TCP Protocol

Exercises TCP/IP Networking With Solutions

Practice Problems: P22, 23, P24, P25, P26, P27, P28, P29, P30, P32, P44 and P45. Hand-in the following: P27, P28, P32, P37, P44

CS321: Computer Networks Congestion Control in TCP

EECS 122. University of California Berkeley

EECS 122. University of California Berkeley. Network Architecture Network hierarchy Layering Performance. Link Layer Ethernet Wi-Fi

FACULTY OF COMPUTING AND INFORMATICS

Error Detection Codes. Error Detection. Two Dimensional Parity. Internet Checksum Algorithm. Cyclic Redundancy Check.

ECE 697J Advanced Topics in Computer Networks

Transmission Control Protocol. ITS 413 Internet Technologies and Applications

Data Link Control Protocols

Systems. Roland Kammerer. 10. November Institute of Computer Engineering Vienna University of Technology. Communication Protocols for Embedded

TCP Congestion Control

FINAL May 21, minutes

ETSN01 Exam. August 22nd am 1pm. Clearly label each page you hand in with your name and the page number in the bottom right hand corner.

Page 1. Goals for Today" Discussion" Example: Reliable File Transfer" CS162 Operating Systems and Systems Programming Lecture 11

EECS489 Computer Networks, Take-home Makeup Midterm (Winter 2007) due in class Wednesday 3/28

PROBLEMSAND EXERCISES

3. Examples. Contents. Classical model for telephone traffic (1) Classical model for telephone traffic (2)

CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS

Announcements. TAs office hours: Mohamed Grissa: Mohamed Alkalbani:

Congestion Control for High Bandwidth-delay Product Networks

Congestion Control in TCP

EECS 122, Lecture 19. Reliable Delivery. An Example. Improving over Stop & Wait. Picture of Go-back-n/Sliding Window. Send Window Maintenance

CSCE 463/612 Networks and Distributed Processing Spring 2017

Studying Fairness of TCP Variants and UDP Traffic

Reasons not to Parallelize TCP Connections for Fast Long-Distance Networks

A Generalization of a TCP Model: Multiple Source-Destination Case. with Arbitrary LAN as the Access Network

Computer Networks Principles

Network Performance: Queuing

Congestion Control In the Network

Chapter 3 Review Questions

Congestion Control for High Bandwidth-delay Product Networks. Dina Katabi, Mark Handley, Charlie Rohrs

Multiple Access Channels

TCP Congestion Control

College of Computer and Information Sciences Department of Computer Engineering CEN444 Computer Networks Midterm 2 Exam Second Semester 1434/1435

There are 10 questions in total. Please write your SID on each page.

CS321: Computer Networks Introduction to Computer Networks and Internet

Transcription:

Models Motivation Timing Diagrams Metrics Evaluation Techniques TOC Models

Motivation Understanding Network Behavior Improving Protocols Verifying Correctness of Implementation Detecting Faults Choosing Provider Feasibility of Applications Monitoring Service Level Agreements Billing TOC Models Motivation

Timing Sending one packet Illustration Examples Queuing Queuing example Store and forward S&F:Multiple packets Cut-Through Fluid View TOC Models Timing

One Packet P bits R bps T seconds T Time P/R P/R = Transmission time T = Propagation time = L/speed = L(km)x(time/km) Time/km = 3.3µs in free space 4 µs in copper 5 µs in fiber TOC Models Timing One Packet

Illustration Transmission Line Transmitter Receiver Signal One bit Each bit takes 1/R seconds to be transmitted The bits take T seconds to propagate TOC Models Timing Illustration

Examples T >> P/R T P = 1KByte R = 1Gbps 100km, fiber T = 500µs P/R = 8µs T << P/R T P/R P/R P = 1KByte R = 100Mbps 1km, fiber T = 5µs P/R = 80µs TOC Models Timing Examples

Queuing Link: P bits Q R bps T seconds Q/R Q/R = queuing delay (load-dependent) T P/R Time TOC Models Timing Queuing

Queuing Example R bps Link: P bits Q 1-kbit packets; R = 1Mbps 1ms Time T(t) = Q/R + P/R for a packet that arrives at t 1ms Time t Let Tn = Q/R +P/R for packet n T1 = 1ms T2 = 1.5ms T3 = 2ms T4 = 1ms T5 = 1.5ms T6 = 1ms TOC Models Timing Queuing Example

Store and Forward System: 10Mbps 5Mbps 100Mbps 10Mbps TOC Models Timing Store and Forward

S&F - Multiple System: 10Mbps 5Mbps 100Mbps 10Mbps TOC Models Timing S&F Multiple

Cut-Through System: R1 = 10Mbps R2 = 10Mbps Header Start forwarding as soon as the header is received Note: What if R2 > R1? TOC Models Timing Cut-Through

A Fluid View System: Rate R A(t); rate a(t) X(t) D(t); rate d(t) # bits in [0, t] # bits in [0, t] a(t) d(t) X(t) TOC Models Timing Fluid View

Metrics Throughput Delay TOC Models Metrics

Throughput Definitions Example 1: Connection Example 2: Link Fluctuations Measurements TOC Models Metrics Throughput

Definitions Roughly: throughput = bit rate (e.g., 120Kbits/second) More precisely: Throughput of a connection or of a link: total number of bits during some period [t, t + T] divided by T Bandwidth* of a link = link rate = bit rate of the transmitter *Note: misnomer, but common usage Utilization of a link = throughput of the link / link rate Bit rate units: 1Kbps = 10 3 bps, 1Mbps = 10 6 bps, 1Gbps = 10 9 bps [For memory: 1Kbytes = 2 10 bytes = 1,024 bytes; 1MBytes = 2 20 bytes] Some rates are expressed in packet per second (pps) relevant when the bottleneck is the header processing TOC Models Metrics Throughput Definitions

Connection Connection: Send W bits (window size) Wait for ACKs Repeat Assume that the round-trip time is RTT seconds Throughput = W/RTT bps Source RTT K Destination Numerical Example: W = 64KBytes = 512 kbits = 512x1,024 = 524,288 bits RTT = 200ms Throughput = W/T = 2.6Mbps RTT Time K TOC Models Metrics Throughput Connection

Link 1Mbps link sends 1,000-bit packets Mbps Rate every µs 1 Rate every 40 ms 0.5 Utilization = 50% TOC Models Metrics Throughput Link Time t

Fluctuations Rate varies over time Throughput over last T seconds mean max min Time t TOC Models Metrics Throughput Fluctuations

Measurements TCP: Keep track of number of bytes received Let R(t) = number of bits in [0, t] Throughput over last T seconds = [R(t) R(t T)]/T Link: Device has counter with number of bytes received; calculate as above TOC Models Metrics Throughput Measurements

Delay Definitions Illustration 1 Illustration 2 Little s Result Measurements: Example 1 Measurements: Example 2 TOC Models Metrics Delay

Definitions Delay/Latency of bit (packet, file) from A to B Jitter: The time required for bit (packet, file) to go from A to B Variability in delay Round-Trip Time (RTT) Two-way delay from sender to receiver and back Bandwidth-Delay Product Product of bw and delay, indicates storage capacity of network TOC Models Metrics Delay Definitions

Illustration 1 S 1 2 D Latest bit seen by time t n at point 1 at point 2 Delay of bit n TOC Models Metrics Delay Illustration 1

Illustration 2 S 1 2 D 1 Packet arrival times at 1 20 ms 2 Packet arrival times at 2 Max delay = 100 ms Min delay = 40 ms Jitter = 60 ms TOC Models Metrics Delay Illustration 2

N S Little s Result 1 2 D X(t) T(N) T(N - 1) S = area S = T(1) + + T(N) = integral of X(t) 1 T(1) + + T(N) N X(t)dt = S =. T T N T T Average occupancy = (average delay)x(average arrival rate) TOC Models Metrics Delay Little s Result

Measurements 1 A Good Epoch TOC Models Metrics Delay Measurements 1

Measurements 2 A Worse Epoch TOC Models Metrics Delay Measurements 2

Evaluation Techniques Models + Analysis Models + Simulations Measurements TOC Models Evaluation

Analysis Example: M/M/1 Queue λ µ Arrivals are Poisson with rate λ Service times are exponentially distributed with mean 1/µ Average delay per packet T = 1/(µ λ) = (1/µ)/(1 ρ) where ρ = λ/µ = utilization For instance, 1/µ = 1ms and ρ = 80% => Q = 5ms TOC Models Evaluation Analysis

Simulation Model of traffic Model of routers, links,. Simulation: Time Driven: X(n) = state at time nε X(n+1) = f(x(n), event at time nε) Event Driven: Y(n) = state after event n E(n) = n-th event T(n) = time when event n occurs [Y(n+1), T(n+1)] = g(y(n), T(n), E(n)] Key Step: Output analysis (estimates, confidence intervals.) TOC Models Evaluation Simulation

Measurements Design Experiment Representative? Output Analysis TOC Models Evaluation Measurements