notes03 - Joseph Kabara TELCOM 2120 Department of...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
1 TELCOM 2120 Collection and Analysis of Measurements Joseph Kabara Department of Information Science and Telecommunications University of Pittsburgh TELCOM 2120: Network Performance 2 Spring 2008 Basic technique for performance analysis Basic technique for performance analysis ¾ Run experiments/observe data on a network or prototype ¾ Benchmark performance Strengths ¾ Saleability, accuracy, includes details Weaknesses ¾ Must have equipment (hardware/software) ¾ Difficult to consider all cases/parameter values ¾ Large of amounts of data ¾ Statistical analysis of data ¾ Time consuming Measurements TELCOM 2120: Network Performance 3 Spring 2008 Four basic types of measurement studies Four basic types of measurement studies 1. Statistical Characterization of metric 1. Statistical Characterization of metric ¾ Determine: call blocking at telephone switch, utilization of network links, bit error rate on link, etc. 2. Compare alternate system designs, equipment or parameter settings ¾ Benchmark routers from different vendors, compare different token rotation times in FDDI ring, compare RAS, etc. Types of Measurement Studies Types of Measurement Studies
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 TELCOM 2120: Network Performance 4 Spring 2008 3. Predicting System Performance ¾ Throughput and Delay increase if add 5 workstations to an Ethernet switch. ¾ End-to-end delay of virtual private network link in ISP backbone networks. 4. Parameterise analytical or simulation model or driving a simulation model ¾ Determining mean call holding times of phone calls for use in queueing model. ¾ Gather data to use as traffic source in computer simulation model of alternate network topologies. Types of Measurement Studies TELCOM 2120: Network Performance 5 Spring 2008 Ethernet Sniffer Collecting Measurements Monitors ¾ Active – inject traffic into the network and see it inject traffic into the network and see it ’s behavior s behavior ¾ Passive – set and listen to network traffic set and listen to network traffic Hardware vs. Software TELCOM 2120: Network Performance 6 Spring 2008 Screen Shot of Screen Shot of Sniffer Sniffer
Background image of page 2
3 TELCOM 2120: Network Performance 7 Spring 2008 Sniffer Data TELCOM 2120: Network Performance 8 Spring 2008 Histogram of ping data TELCOM 2120: Network Performance 9 Spring 2008 TCP Dump data TCP Dump data
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 TELCOM 2120: Network Performance 10 Spring 2008 Statistical Characterization of metric ¾ Determine: call blocking at telephone switch, utilization of network links, bit error rate on link, etc. Gather sample data - n data points n data points x 1 , x 2 , … x n Most statistical inference assumes data is independent ¾ First step is check independence with scatter plot or correlation plot Measurement Analysis Measurement Analysis TELCOM 2120: Network Performance 11 Spring 2008 = + = ) ( ) ( ) ( ) ( 1 ) ( 1 n S n X x n S n X x j n n j i n i i j ρ Statistical tool used to describe how much one variable is linearly related to another ¾ Strength of the relationship ¾
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 20

notes03 - Joseph Kabara TELCOM 2120 Department of...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online