As part of a broader organisational restructure, data networking research at Swinburne University of Technology has moved from the Centre for Advanced Internet Architecture (CAIA) to the Internet For Things (I4T) Research Lab.

Although CAIA no longer exists, this website reflects CAIA's activities and outputs between March 2002 and February 2017, and is being maintained as a service to the broader data networking research community.


This page is part of the GENIUS project.

Quake 3 Single Client Traffic Results

Date: 07/10/02
Duration: approximately 16 mins

The following graphs detail some traffic traces conducted with a single client on the "Free for all" server (gs.caia.swin.edu.au) on port 27960. This server has a 5 map rotation (each map lasting 5 minutes) with two bots present. A controlled test sequence was conducted to become familiar with the various in-game and inter-game traffic patterns from the recorded tracefiles.


Figure 1.

The test sequence was as follows:

  1. Log in to server
  2. Active play on the 1st map shooting bots (A)
  3. Change of map (B)
  4. Idle Client on the 2nd map (C)
  5. Change of map (D)
  6. Active play on the 3rd map shooting bots (E)
  7. Log off from server
Inter-packet Arrival Results
Figure 2.

Figure 3.

Figures 2 and 3 reflect the fact that inter-packet intervals have a mean of 50 msec in the Server to Client direction. It is interesting to note the behaviour of histograms IH 6, IH 9, CIH 6 and CIH 9. Here we see that these histograms have a greater distribution of packets arriving nearer to the mean arrival time. This behaviour identifies when the server was changing maps.

Figure 4.

Figure 5.

Figures 4 and 5 reflect the fact that inter-packet intervals have a mean of 25 msec in the Client to Server direction. Also note how for histograms 3 to 8 and 16 to 21 that the packet arrivals are scattered about the mean arrival time - identifying active game play on the first and third maps respectively. In comparison, histograms 10 to 15 are sharply centred about the mean representing an idle client for the second map played.

Packet Length Results
Figure 6.

Figure 7.

The length and cumulative length histograms above (Figures 6 and 7 respectively) illustrate a mean packet length for Server to Client traffic of 100 bytes. For histograms 5 and 9 you will note there is less spread of the packet lengths, thus indicating a change of maps.

Figure 8.

Figure 9.

For Client to Server traffic (Figures 8 - 9), the mean packet length is approximately 60 bytes. Note how there is less scattering of the packet lengths for histograms 10 to 15 - representing an idle client.

In all the above graphs you will observe that almost twice as many histograms have been recorded for the Client to Server traffic flow in comparison to the Server to Client traffic flow. This is due to the greater packet per second rate exhibited by the Client (40 packets per second) as opposed to the Server (20 packets per second) and that histograms are generated every 2000 packets (rather than set time intervals).



 

Swinburne Homepage Site Map Search Index
 

© Swinburne Copyright and disclaimer Privacy Feedback

Last Updated: Wednesday 18-Dec-2002 07:41:04 AEDT
URL:
Maintained by: Grenville Armitage garmitage@groupwise.swin.edu.au
Authorised by: Grenville Armitage garmitage@swin.edu.au

IndexSearchSite MapSwinburne Home Page