QSIM Application
Network Visualization (NV) Workshop
The SAS/OR Network Visualization (NV) Workshop is an interactive, graphics-oriented tool for investigating network or graph data. Network data consists of a collection of nodes and a series of links between the nodes. Both nodes and links typically have data values associated with them representing the details of their particular application. For example, in a communications network problem the nodes might represent switches or routers and the information associated with each node might include capacity values, device type, traffic volume data, number of dropped packets, etc. The links in this case might represent transmission facilities or media connecting the various nodes and the data associated with the links might be failure rates, error rates, traffic volume, etc.
Input
The network information is described by the following data sets:
-
The link data set contains information on node connectivity and link attributes.
- The node data set contains information on node attributes.
Output
NV Workshop generates two categories of plots
-
Network plot - a visualization of the network data, with options to variety of different node layout
algorithms to obtain alternative visualizations of the same data
- Statistical plots - include histogram, boxplot, and scatterplot graph, to facilitate basic exploratory data analysis.
The data table feature in NV Workshop provides spreadsheet-like, row and column views of the actual data values, while the plot facilities can be used to produce graphic renderings of the data. All plots and data tables are dynamically linked so that selections in one display are automatically reflected in all other displays. For example, selecting bars in a histogram created from node data will automatically highlight the nodes corresponding to those bars in any other plots as well.
In addition to the data table and plot tools, NV Workshop provides facilities for assigning colors, shapes, and labels to data observations. These colors and shapes are then used in the plots when individual points are displayed. These techniques have also proven to be very effective for identifying patterns in the data.
Statistics and Operations Research Home Page | Interactive Visualization and Simulation