August Log Book Entries

<=July September=>

2/8

Delay generator super class finalized now after some furious debugging. Going to implement all the delay/loss models now. This is really simple as it just involves sticking distributions into the generators. Since most of these distributions are pre-made classes from the JNL mathematical library, this should be easy.

4/8

Finally finished all the implementation and testing of all the different types of delay generators. Actually tested them by running them with the actual TCP modules implemented so far. Everything seems okay and I do not see any huge problems. Going to see Jim soon about what I have done.

7/8

Jim seems pleased with my progress. Baring further testing, I am ready to start working on the GUI for the Simulator. Going to do the configuration windows first as they are pure drudge work and I need to practice my Java GUI programming skills again.

12/8

Finished most of the configuration windows. Had some problems with the TCP Layer config window as I had to deal with the work of loading in patch maps but it all works now. Going to start drawing up the main GUI next.

14/8

I have came up with a sample main GUI design. Saw Jim about it today and conferred with him about what he expects to see on the main GUI. What he wants should not be a problem as I know in theory how everything he wants can be implemented. Or at least I HOPE so. The strip graphs displaying the delay/loss values may be a problem.

20/8

I knew it.

The damn strip graphs are a problem. I tried loading in some public domain Java Strip graphs implemented by other people but they simply do not display delay/loss in a manner I need. I will have to implement all this stuff myself. So much for the easy way out. Time to go read up on how to draw on panels in Java.

21/8

I think I have the basics of drawing and I kinda know how strip graphs can be implemented now. Going to start coding tonight on it.

23/8

Finished coding the delay/loss strip graphs. Implementing it into the main GUI tonight with all the other GUI components. Hope this stuff works.

24/8

It works. The main GUI is updating as required as the TCP Simulator runs. I am going to show Jim this at our next meeting so he can tell me whether this is what he envisaged. I am holding off on the panel that actually displays the segments though, as this would be much more work then all the other components as it is such an important part of the GUI.

28/8

Jim seems okay with all the GUI stuff... except for the strip graphs. (Argh) He wants the strip graphs to display only the delay/loss experienced by segments, and not all the overall delay/loss values generated. Oh well, this shouldn't be difficult, I just have to stick in a different set of values into the class that draws the strip graph.