estadium Project Lab 2: Iperf Command

Similar documents
Network Test and Monitoring Tools

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

ECE 697J Advanced Topics in Computer Networks

CHAPTER 3 GRID MONITORING AND RESOURCE SELECTION

ETSF10 Internet Protocols Transport Layer Protocols

Traffic Generator for IP Networks (IPv4 & IPv6) FTTx, LAN, MAN, WAN, WLAN, WWAN, Mobile, Satellite, PLC, etc.

PC BASED REAL TIME DATA EXCHANGE ON 10GbE OPTICAL NETWORK USING RTOS*

46PaQ. Dimitris Miras, Saleem Bhatti, Peter Kirstein Networks Research Group Computer Science UCL. 46PaQ AHM 2005 UKLIGHT Workshop, 19 Sep

Transport Layer PREPARED BY AHMED ABDEL-RAOUF

Washington State University CptS 455 Sample Final Exam (corrected 12/11/2011 to say open notes) A B C

Southern Polytechnic State University Spring Semester 2009

Models. Motivation Timing Diagrams Metrics Evaluation Techniques. TOC Models

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

Mobile Transport Layer Lesson 02 TCP Data Stream and Data Delivery

Dropping Packets in Ubuntu Linux using tc and iptables

Performance and Bandwidth Testing for Data Circuits

Introduction to Networking and Systems Measurements

Objective To examine the throughput of a TCP connection as the flow control window size is varied.

Comparision study of MobileIPv4 and MobileIPv6

ECE697AA Lecture 3. Today s lecture

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

II. Principles of Computer Communications Network and Transport Layer

CSE 123: Computer Networks

Computer Networks Project 4. By Eric Wasserman and Ji Hoon Baik

Internet Tool Practice. 이지민 장동현

Experiments on TCP Re-Ordering March 27 th 2017

Delay Controlled Elephant Flow Rerouting in Software Defined Network

Equation-based Congestion Control

Layer 4 TCP Performance in carrier transport networks Throughput is not always equal to bandwidth

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.

Review Questions for Midterm Exam-2 Fall 2016

Examen final de Xarxes de Computadors (XC) 6/6/2011 NAME: SURNAME DNI:

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

Fermilab WAN Performance Analysis Methodology. Wenji Wu, Phil DeMar, Matt Crawford ESCC/Internet2 Joint Techs July 23, 2008

xkcd.com End To End Protocols End to End Protocols This section is about Process to Process communications.

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

SCHOOL OF INFORMATION, COMPUTER AND COMMUNICATION TECHNOLOGY SIRINDHORN INTERNATIONAL INSTITUTE OF TECHNOLOGY THAMMASAT UNIVERSITY ASSIGNMENT REPORT

Brief Overview and Background

An Empirical Characterization of Cellular Network Performance

Your Name: Your student ID number:

Network softwarization Lab session 2: OS Virtualization Networking

Congestion Avoidance and Control. Rohan Tabish and Zane Ma

The Transmission Control Protocol (TCP)

521262S Computer Networks 2 (fall 2007) Laboratory exercise #4: Multimedia, QoS and testing

Exercises TCP/IP Networking With Solutions

EECS 3214: Computer Network Protocols and Applications. Final Examination. Department of Computer Science and Engineering

Research Article A Novel Solution based on NAT Traversal for High-speed Accessing the Campus Network from the Public Network

Computer Networks. Homework #4: No Grading

Ashortage of IPv4 address space has

CS 638 Lab 6: Transport Control Protocol (TCP)

15-441: Computer Networks Homework 3

Transport Protocols and TCP

Congestion Control. COSC 6590 Week 2 Presentation By Arjun Chopra, Kashif Ali and Mark Obsniuk

LANCOM Techpaper IEEE n Indoor Performance

Network Management & Monitoring

First Exam for ECE671 Spring /22/18

Problem Set 7 Due: Start of Class, November 2

Student ID: CS457: Computer Networking Date: 3/20/2007 Name:

Introduction to Troubleshooting TCP/IP Networks with Wireshark

Performance investigation and comparison between virtual networks and physical networks based on Sea-Cloud Innovation Environment

Performance Modeling

cs/ee 143 Fall

6.1 Internet Transport Layer Architecture 6.2 UDP (User Datagram Protocol) 6.3 TCP (Transmission Control Protocol) 6. Transport Layer 6-1

2.993: Principles of Internet Computing Quiz 1. Network

EXPERIMENTAL STUDY OF FLOOD TYPE DISTRIBUTED DENIAL-OF- SERVICE ATTACK IN SOFTWARE DEFINED NETWORKING (SDN) BASED ON FLOW BEHAVIORS

Streaming Video and TCP-Friendly Congestion Control

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

Transport Layer. <protocol, local-addr,local-port,foreign-addr,foreign-port> ϒ Client uses ephemeral ports /10 Joseph Cordina 2005

Estimation of available bandwidth

ICS 451: Today's plan. Sliding Window Reliable Transmission Acknowledgements Windows and Bandwidth-Delay Product Retransmission Timers Connections

ATL : An Adaptive Transport Layer Protocol Suite for Next Generation Wireless Internet

EXAM TCP/IP NETWORKING Duration: 3 hours With Solutions

Performance Evaluation of NEMO Basic Support Implementations

EC441 Midterm Two Fall 2017

TIXway IRT Lab Evaluation

CS321: Computer Networks Congestion Control in TCP

Midterm Review EECS 122. University of California Berkeley

EXAM TCP/IP NETWORKING Duration: 3 hours

Exercise 1, Basic Network Measurements. Task 1. Latency for cctld name servers. Maryam Tavakoli Momtaz

Lab Exercise UDP & TCP

Performance of Host Identity Protocol on Lightweight Hardware

Monitoring in Disadvantaged Grids

Version 2.6. Product Overview

Advanced Computer Networks

Traffic Generator for IP Networks (IPv4 & IPv6) FTTx, LAN, MAN, WAN, WLAN, WWAN, Mobile, Satellite, PLC, etc.

Broadband Quality of Service

CS 326: Operating Systems. Networking. Lecture 17

Advanced Computer Networks

EXAM TCP/IP NETWORKING Duration: 3 hours

A common issue that affects the QoS of packetized audio is jitter. Voice data requires a constant packet interarrival rate at receivers to convert

PfR Voice Traffic Optimization Using Active Probes

Performance Analysis of the Intertwined Effects between Network Layers for g Transmissions

Tutorial 8 : Congestion Control

TCP Throughput Testing

A Policy Controlled IPv4/IPv6 Network Emulation Environment

SCTP: A new networking protocol for super-computing. Mohammed Atiquzzaman Shaojian Fu Department of Computer Science University of Oklahoma.

HP 830 Series PoE+ Unified Wired-WLAN Switch Switching Engine

Congestion control mechanism of TCP for achieving predictable throughput

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

Proxy-based TCP-friendly streaming over mobile networks

Transcription:

estadium Project Lab 2: Iperf Command Objectives Being familiar with the command iperf. In this Lab, we will set up two computers (PC1 and PC2) as an ad-hoc network and use the command iperf to measure network parameters of TCP (Transmission Control Protocol) connection and UDP (User Datagram Protocol) connection. Mainly, we use iperf on TCP connection to measure the TCP throughput (Part B), which will lead to calculate the TCP window size. On the other hand, we use iperf on UDP connection to measure UDP throughput, jitter, and data loss (Part C). In addition, there are some iperf s options that we will take a look (Part D G). Remark: One PC will act as a server (=receiver). Another PC will act as a client (=transmitter). The recorded measurement values are the values shown on the server, not the client. The setup with more than two PCs is possible. To install iperf, on Ubuntu, type >> sudo apt-get install iperf Note: This will install iperf online, then make sure that the internet must be connected during the installation. Please see the references for comprehensive explanation and examples. Please see the references or type man iperf to see the iperf command s options. If not specified, just put two PCs apart with a convenient distance (1-2m). A: Setup the Ad-hoc Network A.1) Set up the ad-hoc network between two PCs by following the steps in Lab 1. Note that let the IP address of PC1 is PC1_IPaddress (for example, 192.168.12.1). B: TCP Connection TCP Throughput Measurement B.1) Set PC1 as a server (receiver) by typing B.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress The screens are shown below

Remark: The server must be set up first. The measured TCP throughput is 4.92 Mbit/s (the throughput shown on the server s screen). By default, iperf runs a 10 second test. By default, TCP window is 85.3 KByte at the server. By default, TCP window is 16.0 KByte at the client. When we know the TCP throughput, we can calculate an appropriate TCP window size (of the client) following this equation: TCP window size = (Throughput) X (Round Trip Time) where the round trip time (RTT) is measured by using ping command. C: UDP Connection UDP Throughput, Jitter, and Loss Measurement C.1) Set PC1 as a server (receiver) by typing -u C.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress -u The client will send the UDP packets with bandwidth 1 Mb/s (default). The screens are shown below

Remark: The server must be set up first. The measured UDP throughput is the throughput shown on the server s screen. By default the UDP throughput is around 1 Mbit/s. The measured UDP jitter is shown on the server s screen, which is 1.033 ms. The measured data loss is shown on the server s screen, which is 0. By default, iperf runs a 10 second test. By default, UDP buffer is 110 KByte at the server. By default, UDP buffer is 110 KByte at the client. D: Iperf s Option - TCP Connection with a Varying Observed Time Duration D.1) Set PC1 as a server (receiver) by typing D.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress -t 5 The client will send TCP packets for 5 second. Note that this option is also available for UDP connection. The screens are shown below. E: Iperf s Option - TCP Connection with a Varying Observed Time Interval

E.1) Set PC1 as a server (receiver) by typing E.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress -i 2 The client will send TCP packets for 10 second (default) and show the details every 2 second. Note that this option is also available for UDP connection. The screens are shown below. F: Iperf s Option - TCP Connection with a Varying TCP Window Size F.1) Set PC1 as a server (receiver) by typing F.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress -w 150k The client will send TCP packets for 10 second (default) with requesting the TCP window size 150 KByte. Note that this option is also available for UDP connection but with a different meaning. The screens are shown below.

G: Iperf s Option - UDP Connection with a Varying Bandwidth G.1) Set PC1 as a server (receiver) by typing -u G.2) Set PC2 as a client (transmitter) by typing >> iperf -c PC1_IPaddress -u -b 5M The client will send UDP packets for 10 second (default) with the bandwidth 5 Mbit/s. The screens are shown below. References: [1] http://openmaniak.com/iperf.php [2] http://pirlwww.lpl.arizona.edu/resources/guide/software/iperf/ [3] http://kb.pert.geant2.net/pertkb/iperftool