PingER (Ping end-to-end ) at PERN2This technical report outlines the project PingER and its deployment at various PERN POP...
1|PageTable of Contents1.   INTRODUCTION ....................................................................................
2|Page                                                       Table of FiguresFigure 1: PingER worldwide coverage (Google M...
3|Page1. INTRODUCTIONThe Internet End-to-end Performance Measurement (IEPM) group at the Stanford LinearAccelerator Center...
4|Page3. INSTALLATION & CONFIGURATIONFollowing are the installation requirements for PingER monitoring machines;         ...
5|Page          o Chmod a+x traceroute.pl      Test it by loading the URL http://yourwebbrowser/cgi-bin/traceroute.pl   ...
6|Page                      Figure 2: PingER Pakistan coverage (Google Map)PingER is deployed at following PERN2 POPs;    ...
7|Page6. RESULTS OF PINGER MONITORING AT PERN POPS     RTT (Round Trip Time):      Round-trip time (RTT) depends on the d...
8|Page                                        RTT monitored from HEC KHI POP                          35                  ...
9|Page                                          RTT monitored from HEC PSH POP                          60                ...
10 | P a g e                                                         RTT monitored from QAU POP                           ...
11 | P a g e                                                        RTT monitored from GCU POP                            ...
12 | P a g e       Packet Loss:        Packet Loss Metrics Standard Values are given in table below:        Packet Loss  ...
13 | P a g e                         Packet loss as monitored from HEC KHI POP                   2.5                     2...
14 | P a g e                          Packet loss as monitored from HEC PSH POP                    4                    3 ...
15 | P a g e                               Packet loss as monitored from QAU POP                      3                   ...
16 | P a g e                          Packet loss as monitored from GCU POP                   2.5                     2   ...
17 | P a g e7. REFERENCES[1] PingER official website http://www-iepm.slac.stanford.edu/pinger/[2] PingER installation http...
Upcoming SlideShare
Loading in …5
×

Technical report pinger

390 views

Published on

SLAC

Published in: Business, Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
390
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Technical report pinger

  1. 1. PingER (Ping end-to-end ) at PERN2This technical report outlines the project PingER and its deployment at various PERN POPs. Themost recent data collected from PingER nodes deployed at PERN2 POPs is also included. 4/30/2012
  2. 2. 1|PageTable of Contents1. INTRODUCTION ..................................................................................................................................... 32. WORKING MECHANISM ........................................................................................................................ 33. INSTALLATION & CONFIGURATION....................................................................................................... 44. USES OF THE PINGER DATA................................................................................................................... 55. PINGER DEPLOYED AT PERN POPS ........................................................................................................ 56. RESULTS OF PINGER MONITORING AT PERN POPS .............................................................................. 77. REFERENCES ........................................................................................................................................ 17
  3. 3. 2|Page Table of FiguresFigure 1: PingER worldwide coverage (Google Map).................................................................................... 3Figure 2: PingER Pakistan coverage (Google Map) ....................................................................................... 6Figure 3: Round Trip Time as monitored from HEC Islamabad POP ............................................................. 7Figure 4: Round Trip Time as monitored from Faisalabad POP .................................................................... 7Figure 5: Round Trip Time as monitored from HEC Karachi ......................................................................... 8Figure 6: Round Trip Time as monitored from HEC Lahore POP .................................................................. 8Figure 7: Round Trip Time as monitored from Multan POP ......................................................................... 8Figure 8: Round Trip Time as monitored from HEC Peshawar ..................................................................... 9Figure 9: Round Trip Time as monitored from Hyderabad POP ................................................................... 9Figure 10: Round Trip Time as monitored from Fatima Jinnah Women University POP.............................. 9Figure 11: Round Trip Time as monitored from Quaid e Azam University POP ......................................... 10Figure 12: Round Trip Time as monitored from College of Physicians and Surgeons Pakistan POP .......... 10Figure 13: Round Trip Time as monitored from Karachi University POP .................................................... 10Figure 14: Round Trip Time as monitored from Government College University POP .............................. 11Figure 15: Round Trip Time as monitored from Punjab University POP .................................................... 11Figure 16; Round Trip Time as monitored from Quetta POP ...................................................................... 11Figure 17: Packet Loss as monitored from HEC Islamabad ......................................................................... 12Figure 18 Packet Loss as monitored from Faisalabad POP ......................................................................... 12Figure 19: Packet Loss as monitored from HEC Karachi ............................................................................. 13Figure 20 Packet Loss as monitored from HEC Lahore ............................................................................... 13Figure 21: Packet Loss as monitored from Multan POP ............................................................................. 13Figure 22: Packet Loss as monitored from HEC Peshawar POP .................................................................. 14Figure 23: Packet Loss as monitored from Hyderabad POP ....................................................................... 14Figure 24: Packet Loss as monitored from Fatima Jinnah Women University POP .................................... 14Figure 25: Packet Loss as monitored from Quaid e Azam University POP ................................................. 15Figure 26: Packet Loss as monitored from College of Physicians and Surgeons Pakistan POP .................. 15Figure 27: Packet Loss as monitored from Karachi University POP ............................................................ 15Figure 28: Packet Loss as monitored from Government College University POP....................................... 16Figure 29: Packet Loss as monitored from Punjab University POP............................................................. 16Figure 30: Packet Loss as monitored from Quetta POP .............................................................................. 16
  4. 4. 3|Page1. INTRODUCTIONThe Internet End-to-end Performance Measurement (IEPM) group at the Stanford LinearAccelerator Center (SLAC) initiated their project named PingER for network performancemonitoring. It is led by SLAC and development work is carried from NUST/SEECS, FNAL, andICTP/Trieste. Earlier, in 1995 it was for the High Energy Physics community. However, thiscentury it has more focus on measuring the Digital Divide from an Internet Performanceviewpoint. The project now involves measurements to over 700 sites in over 160 countries [1]. Figure 1: PingER worldwide coverage (Google Map)The PingER infrastructure reflects the wide geographical spread of the collaborations, andinvolves a large number of research & education networks and commercial networks.2. WORKING MECHANISMThe framework of the PingER project is based on the PING (widely used for networktroubleshooting). A PING involves sending an Internet Control Messages Protocol (ICMP) echorequest to a specified remote node which responds with an ICMP echo reply.PingER measures following Network Key Performance Indicators (KPI);  Packet Loss  Response Time (RTT)  TCP Throughput
  5. 5. 4|Page3. INSTALLATION & CONFIGURATIONFollowing are the installation requirements for PingER monitoring machines;  Hardware: 500MHz or higher, RAM atleast 128MB, 2GB of disk space, 10/100Mbps NIC should be quite adequate  Operating System: LINUX having web service (example CENTOS)  PingER machine must be registered domain name service (DNS).  Uninterrupted power supply (UPS).In addition, firewall of the machine should be enabled, ICMP echo and reply must not beblocked, and ssh access should be enabled for remote access of the machine.Configuration might be done by PingER representative on giving them remote access.Configuration procedure is done in following two steps;Step 1:  On your machine download latest version of PingER from http://pinger.seecs.edu.pk/install1.html  To make the files, type: tar xzf pinger-2.0.2.tar.gz cdpinger-2.0.2 ./configure make test-prereqs make  Make install  Make install_cron  README file for full details.Step2:  Install ping_data.pl  Install a Traceroute Server  Get ping_data.pl and put it in the CGI directory for your HTTP server(often/usr/local/cgi- bin or /usr/lib/cgi-bin)  You may already have cgi-lib.pl in your CGI directory. If not, put it there now  Make ping_data.pl and cgi-lib.pl executable o chmod a+x ping_data.pl cgi-lib.pl  Put traceroute.pl in your CGI directory (often /usr/local/cgi-bin or /usr/lib/cgi-bin)  Make it execitable with the command
  6. 6. 5|Page o Chmod a+x traceroute.pl  Test it by loading the URL http://yourwebbrowser/cgi-bin/traceroute.pl  Open file pinger.xml and do the following changes; o <SrcName>dns name e.g pinger.seecs.edu.pk </SrcName> o <HttpGetBin>Path where lynx installed </HttpGetBin>Example for PingER machine installation and configuration can be found at [2]4. USES OF THE PINGER DATAThe PingER monitored data from a site is used in many different ways. Some of them are asfollows; 1. Selecting Internet Service Provider. When someone buys a network connection, he/she should be able to check if the quality of the connection is satisfactory. Using the PingER data, one can check if they are getting what they are paying for in terms of link quality, throughput etc. 2. Trouble Shooting Tool. PingER has been used to determine if a reported problem is related to networks, identify the time at which the problem has started, decide whether it is still occurring, and provide quantitative analysis for Internet Service Providers (ISPs). 3. Network Performance. The PingER data can be used to select universities from developing countries for remote collaboration programs. By using PingER to measure the Packet Loss and RTT, it is possible to provide expectations on the performance of bulk data transfers, VOIP, video conferencing and other applications. In short, using PingER monitored data one can determine how well any interactive applications might work between various pairs of sites. 4. Digital Divide. The PingER monitored data is used to quantify the difference in performance between developed and developing countries.Some real-time examples of the use of PingER monitored data can be found [3].5. PINGER DEPLOYED AT PERN POPSPingER project is also monitoring Pakistan Education and Research Network (PERN). PingER hasdeployed the monitoring system at PERNs 15 Point of Presence (POPs) nationwide to collectthe Network Key Performance Indicators (KPI).
  7. 7. 6|Page Figure 2: PingER Pakistan coverage (Google Map)PingER is deployed at following PERN2 POPs; 1. Higher Education Commission Islamabad 2. Air University Islamabad 3. Quaid-e-Azam University Islamabad 4. Fatima Jinnah Women University Rawalpindi 5. Higher Education Commission Peshawar 6. Higher Education Commission Lahore 7. Govt. College University Lahore 8. Punjab University Lahore 9. Bahauddin Zakariya University, Multan 10. University of Agriculture Faisalabad 11. Higher Education Commission Karachi 12. University of Karachi 13. College of Physician and Surgeon Pakistan Karachi 14. Sindh University Hyderabad 15. Baluchistan University of Information Technology, Engineering and Management Sciences Quetta
  8. 8. 7|Page6. RESULTS OF PINGER MONITORING AT PERN POPS  RTT (Round Trip Time): Round-trip time (RTT) depends on the distance between the sites plus the delay at each hop along the path between the sites. RTT also depends on traffic load and congestion (packet size). RTT Metrics Standard Values are given in table below: RTT Excellent Good Acceptable Very Poor Bad <62.5 >=62.5 & > =125 & >= 250 & >= 500 <125 <250 <500 RTT monitored from HEC ISB POP 60 milli-second 40 20 0 Figure 3: Round Trip Time as monitored from HEC Islamabad POP RTT from UAF FSB POP 80 milli-second 60 40 20 0 Figure 4: Round Trip Time as monitored from Faisalabad POP
  9. 9. 8|Page RTT monitored from HEC KHI POP 35 30 milli-second 25 20 15 10 5 0 Figure 5: Round Trip Time as monitored from HEC Karachi RTT monitored from HEC LHR POP 150 milli-second 100 50 0 Figure 6: Round Trip Time as monitored from HEC Lahore POP RTT monitored from BZU MTN POP 70 60 milli-second 50 40 30 20 10 0 Figure 7: Round Trip Time as monitored from Multan POP
  10. 10. 9|Page RTT monitored from HEC PSH POP 60 50 milli-second 40 30 20 10 0 HEC FJWU QAU HEC GCU PU MTN FSB HEC CPSP UOK HYD QTA ISB POP POP LHR POP POP POP POP KHI POP POP POP POP POP POP POP Figure 8: Round Trip Time as monitored from HEC Peshawar RTT monitored from HYD POP 140 120 milli-second 100 80 60 40 20 0 Figure 9: Round Trip Time as monitored from Hyderabad POP RTT monitored from FJWU POP 60 50 milli-second 40 30 20 10 0 Figure 10: Round Trip Time as monitored from Fatima Jinnah Women University POP
  11. 11. 10 | P a g e RTT monitored from QAU POP 60 50 milli-second 40 30 20 10 0 Figure 11: Round Trip Time as monitored from Quaid e Azam University POP RTT monitored from CPSP POP 50 milli-second 40 30 20 10 0 Figure 12: Round Trip Time as monitored from College of Physicians and Surgeons Pakistan POP RTT monitored from KU POP 40 milli-second 30 20 10 0 Figure 13: Round Trip Time as monitored from Karachi University POP
  12. 12. 11 | P a g e RTT monitored from GCU POP 150 milli-second 100 50 0 Figure 14: Round Trip Time as monitored from Government College University POP RTT monitored from PU POP 160 140 milli-second 120 100 80 60 40 20 0 Figure 15: Round Trip Time as monitored from Punjab University POP RTT monitored from QTA POP 200 milli-second 150 100 50 0 HEC FJWU QAU HEC HEC GCU PU MTN FSB HEC CPSP UOK HYD ISB POP POP PSH LHR POP POP POP POP KHI POP POP POP POP POP POP POP Figure 16; Round Trip Time as monitored from Quetta POP
  13. 13. 12 | P a g e  Packet Loss: Packet Loss Metrics Standard Values are given in table below: Packet Loss Excellent Good Acceptable Very Poor Bad <0.1% >=0.1% & < > =1% & < >= 5% & < >= 12% 1% 2.5% 12% TCP connection fails when Packet losses greater that 10 % Packet loss as monitored from HEC ISB POP 2.5 2 1.5 % 1 0.5 0 Figure 17: Packet Loss as monitored from HEC Islamabad Packet loss as monitored from FSB POP 4 3.5 3 2.5 2 % 1.5 1 0.5 0 Figure 18 Packet Loss as monitored from Faisalabad POP
  14. 14. 13 | P a g e Packet loss as monitored from HEC KHI POP 2.5 2 % 1.5 1 0.5 0 Figure 19: Packet Loss as monitored from HEC Karachi Packet Loss as monitored from HEC LHR POP 2.5 2 1.5 % 1 0.5 0 Figure 20 Packet Loss as monitored from HEC Lahore Packet loss as monitored from MTN POP 5 4 3 % 2 1 0 Figure 21: Packet Loss as monitored from Multan POP
  15. 15. 14 | P a g e Packet loss as monitored from HEC PSH POP 4 3 2 % 1 0 Figure 22: Packet Loss as monitored from HEC Peshawar POP Packet loss as monitored from HYD POP 2 1.5 1 % 0.5 0 Figure 23: Packet Loss as monitored from Hyderabad POP Packet loss as monitored from FJWU POP 2.5 2 1.5 % 1 0.5 0 Figure 24: Packet Loss as monitored from Fatima Jinnah Women University POP
  16. 16. 15 | P a g e Packet loss as monitored from QAU POP 3 2.5 2 1.5 % 1 0.5 0 Figure 25: Packet Loss as monitored from Quaid e Azam University POP Packet loss as monitored from CPSP POP 2.5 2 1.5 % 1 0.5 0 Figure 26: Packet Loss as monitored from College of Physicians and Surgeons Pakistan POP Packet loss as monitored from KU POP 2.5 2 1.5 % 1 0.5 0 Figure 27: Packet Loss as monitored from Karachi University POP
  17. 17. 16 | P a g e Packet loss as monitored from GCU POP 2.5 2 % 1.5 1 0.5 0 Figure 28: Packet Loss as monitored from Government College University POP Packet loss as monitored from PU POP 2.5 2 1.5 % 1 0.5 0 Figure 29: Packet Loss as monitored from Punjab University POP Packet loss as monitored from QTA POP 2 1.5 1 % 0.5 0 Figure 30: Packet Loss as monitored from Quetta POP
  18. 18. 17 | P a g e7. REFERENCES[1] PingER official website http://www-iepm.slac.stanford.edu/pinger/[2] PingER installation http://www-iepm.slac.stanford.edu/pinger/epinger-install.html[3] PingER uses http://www-iepm.slac.stanford.edu/pinger/uses.html[4] PingER monitored Data http://www-wanmon.slac.stanford.edu/cgi-wrap/pingtable.pl

×