5. Visualizing the results with Plove and Scalars

PREV: 4. Adding statistics collection UP: Contents

Scalar statistics

The Scalars tool can be used to visualize the contents of the omnetpp.sca file. It can draw bar charts, x-y plots (e.g. throughput vs offered load), or export data via the clipboard for more detailed analysis into spreadsheets or other programs.

$ scalars omnetpp.sca

The program displays the data in a table with columns showing the file name, run number, module name where it was recorded, and the value. There're usually too many rows to get an overview, so you can filter by choosing from (or editing) the three combo boxes at the top. (The filters also accept the *, ** wildcards.)

sca1.gif

You could actually load further scalar files into the window, and thus analyse them together.

You can copy the selected rows to the clipboard by Edit|Copy or the corresponding toolbar button, and paste them e.g. into OpenOffice Calc, MS Excel or Gnumeric.

The bar chart toolbar button creates -- well -- a bar chart in a new window. You can customize the chart by right-clicking on it and choosing from the context menu. It can also be exported to EPS, GIF, or as metafile via the Windows clipboard (the latter is not available on Unix of course).

sca2.gif

Plotting the output vectors

Output vector files can be visualized with Plove. Try the following:

$ plove omnetpp.vec

The left pane displays vectors that are present in the omnetpp.vec file. (You can load further vector files as well.)

To plot, you have to copy some vectors to the right pane, select one or more of them (shift+click and ctrl+click works), and click the Plot icon on the toolbar.

plove1.gif

The graph is displayed in a separate window.

plove2.gif

As in Scalars, you can customize the graph and export it in various formats. For the following screenshot we turned off connecting the data points.

plove3.gif

We can apply a filter which plots mean on [0,t). In the main window, right-click the selected vectors, then choose Pre-plot filtering from the context menu.

plove4a.gif

Once in the dialog, choose 'mean' from the filter dropdown list and click OK. Next time you click the Plot button on the toolbar, you'll get the filtered charts.

plove4.gif

Conclusion

Hope you have found this tutorial a useful introduction into OMNeT++. Comments and suggestions will be appreciated.

UP: Contents


Generated on Sat Oct 21 17:48:02 2006 for Tictoc Tutorial by  doxygen 1.4.6