Overview

The simple way to get at the data is not to use the archive (since the data needs cleansing) but to use the form https://www.slac.stanford.edu/cgi-bin/pingtable.pl . This enables you to select the metric, packet size, source (Measurement Agent - MA), the target group, the time aggregation, etc.  There is also a filter which can be used for removing bad data from the monthly and hence yearly data. Bad data is typically from a host that at first looks to be in a certain region (e.g. China) but the RTT as seen from SLAC is impossibly small. 
Once you have the relevant table then cut and paste it into Excel and use that for the manipulation.
There is also a script prm.pl that uses the pingtable.pl form (via wget). This is used to access pingtable.pl multiple times to aggregate the data from an MA to multiple target groups for each metric, by year or by month. 
  • The script prm.pl, is used to create csv files for the analyzed data which can then be used by PingER motion charts or PingER Executive Plots.

  • The script requires a configuration file, which contains entries for all the reports prm should create. These entries are of the form:

     

    <metric name>    <monitoring site>    <country||continent>    <tick>    (<filter>?)
  • It gets its data from pingtable.pl either by calling it locally or by accessing it via the web.
  • The reports created by prm are available online.
  • The reports are used by the metric motion plots, the metric maps and the executive plots.

An example oif the output is below:

allyearly,?,1998,1999,2000,2001,2002,2003,2004,2005,2006,2007,2008,2009,2010

EDU.SLAC.STANFORD.N3-to-Afghanistan,.,.,.,.,.,.,.,.,.,.,12.504,4.481,9.257,5.284

EDU.SLAC.STANFORD.N3-to-Albania,.,.,.,.,.,.,.,.,.,.,.,4.949,7.133,4.590

 later it is atomically copied to files in the permanent directory

How to retrieve data from the SLAC anonymous FTP archive:

Should you decide to go ahead I am happy to help/hold your hand and improve the documentation.

Data Flow

The mechanism of gathering, archiving the data, cronjobs  etc is described in https://confluence.slac.stanford.edu/display/IEPM/PingER+data+flow+at+SLAC . this also describes the raw and analyzed data.

A period(.) means there was no data.

  /afs/slac/g/www/www-iepm/pinger/prmout

There is a directory with lots of these analyses, see https://www-iepm.slac.stanford.edu/pinger/prmout/
  • No labels