Video trace data from the Fairisle ATM network
This trace contains traffic measurements from the AVA-200 ATM Camera
developed by Nemesys Research Ltd, and marketed by Fore Systems.
Specifications for the AVA-200 can be obtained from Nemesys or Fore
Systems (see end of this file). The AVA-200 (also simply called the
ATM Camera) was set up to transmit video over the Fairisle network.
The measurements were made using code written by Simon Crosby and
other members of the Systems Research Group at the University of
This data is Copyright (c) Simon Crosby and Ian Pratt, 1994, University of
Cambridge, UK. You may freely distribute this data but must retain the
Details of the Fairisle network and the Systems Research Group
activities can be obtained from the ftp site at Cambridge:
ftp host is: ftp.cl.cam.ac.uk
This is also available as the Systems Research Group Technical Note of
March 1994 (the blue book), and can be obtained from the technical
reports service at Cambridge: send email to email@example.com
ATM camera traces
ftp host is: ftp.cl.cam.ac.uk
Brief description of the traces in the subdirectories:
first-trace - the old ava trace which I took ages ago (1994)
misc - miscellaneous clips of video
itn = ITN news including interviews and news clips
lethal-weapon-3 - traces from the movie of the same name,
at different peak rates
talk-show - an anglia TV talk show with voting
(so there were phone #'s displayed
on the screen, with audience participation.
In general files are named as follows:
<name>.x.y = <name>.peakrate.fps (jpeg q 20, half field video)
the peak rate here is the number on the nrlwish slider bar.
If a file has a third component to name, that is just an instance number
This data currently comprises a single trace of ATM camera video.
More data will be available soon. The camera was set up to transmit
25 frames per second, JPEG compressed, 24 bits per pixel colour video
from various sources. In this experiment the ATM camera's
spacing policer, which limits the peak emission interval of cells from
the camera, was switched off. Thus the ATM camera transmits cells
whenever they are ready to leave the device, and back-to-back cells
are thus possible. The trace comprises the first 1000000 cells of the
transmission, which included both action scenes (an explosion) and
relatively static portions when credits were scrolling on the screen.
The video data was transmitted from the AVA-200 to a Fairisle port
controller which performed the measurements, and from there to a Sun
Sparc 10 which displayed the video.
The JPEG picture size is 48x33 tiles (each tile is 8x8 pixels). Each
tile is is compressed using a JPEG Q-Factor of 32, and transmitted
with a single cell of tile overhead. The amount of information per
tile will differ according to the video content.
The following is a graph of the trace data:
The file exported for ftp is a gzip format compressed ASCII file.
Use gunzip to uncompress the file and obtain the trace data. If you
do not have access to gzip and gunzip, email Simon.Crosby@cl.cam.ac.uk
and I will try to help you.
The file is in ASCII format, and each line (i>0) is a single measurement.
Each measurement (i) records the time, in ticks, between the absolute
cell number (i) and absolute cell number (i-1). Measurement (1) is
A tick is a fabric cell time on the Fairisle port controller. The
fabric cell time in these experiments was set to 4.2666666 microseconds.
A pair of back to back cells is recorded as an inter-arrival time of 1
The AVA-200 is designed and manufactured by Nemesys Research Ltd in
the UK. e-mail firstname.lastname@example.org. The unit is licensed for
distribution by: K-NET Ltd in Europe. email: email@example.com and
Fore Systems, Inc in the rest of the world. email: firstname.lastname@example.org