From MAGGIE

Contents 1 Project Title 2 Project Aim 3 Motivation 4 Project Description 5 Requirements

[edit] Project Title

Mapping PingER Metrics

[edit] Project Aim

The aim of this project is to provide useful visualization of PingER data in a map form

[edit] Motivation

Currently PingER monitors about 800 hosts in over 100 countries from 30+ monitoring sites in 15 countries around the world. In all this corresponds to about 100,000 measurements per day (or ~2,000,000 pings/day) between over 2200 monitor site/remote site pairs. We need ways to easy visualize all this data to graphically illustrate the differences between various regions both currently and the longer term trends. Such graphs are useful for executives, for network administrators and for Network Operating Centers (NOCs).

[edit] Project Description

Currently PingER is able to present detailed information in tabular format with some drill down to time series plots. It also has the ViPER interactive map based interface to the data being developed by Shahryar Khan and long term trend plots for regions developed by Syed Akbar Mehdi. There is also the Ping World Demo that shows colored regions of the world with time series of pings made once/second for the last 30 minutes (see Figure below).

What we need is a way to use a map as background while displaying PingER data time series in the foreground. In some cases we will need to produce regularly (driven for example by crontab) updated static maps that can be displayed via the web. In other cases there will need to be an interactive front end. The student will need to start out by studying possible ways to display the data.

Possibilities include:

• Use Google Charts API to color countries by the value of the chosen aggregated metric for the country, and/or bar or pie charts of a metric for each country.
• Use Excel/Mapland to color countries by the value of the chosen aggregated metric for the country, or bar or pie charts of a metric for each country. A problem with this may be the lack of a way to automate the updates, or provide interactive access, also there are over 100 countries so the picture will be very busy, and I do not think there is a way to easily define regions;
• Extend PingWorld (it uses Java Analysis Studio) to enable it to get the data from PingER rather than from pinging hosts in realtime. One could extend the background region coloring to shade countries where there are hosts being monitored; (Using Google Charts would be a better idea).
• Extending Shahryar's ViPER: The interactive interface would allow selection of metric, selection of the monitoring host, selection of countries or hosts, selection of regions. Each series for a region could represent a country. Other possibilities could include leveraging some of Syed Akbar's work to add displaying trendline fits as an alternate to individual points, provide for selection of log y scales, anomalous event detection information.

[edit] Requirements

• The student will need to be or become proficient in Unix/Linux, perl, CGI scripts, Java and Javascript.
• The code will need to be production quality. Guidelines on how to write perl will be provided (see IEPM Perl Coding Style and Coding Style). Help on writing clean code is available here and here.
• The student will need to apply for and get a Unix account at SLAC and will be provided access to the relevant computers, files and databases. (contact Umar Kalim for details).