Why would I want to use the OPNET Model Support distribution package instead of the Distribution KPs?

Why would I want to use the OPNET Model Support distribution package instead of the Distribution KPs?

There are many reasons why you want to use the oms_dist_... sub-package over the built-in op_dist_... set of Kernel Procedures. Some of them below:1. The op_dist APIs typically require that you specify the distribution name and distribution arguments (sometimes more than one -- e.g., for normal and uniform distributions). The model writer has the option of providing these input arguments as a single or separate attributes that one can specify. The oms_dist package abstracts all of this by working with the distribution specification user interface and dealing with one simple string argument for specifying the distribution name and arguments. 2. The oms_dist package allows you to use any of the following distribution types without changing the user-written code: - standard, built-in distributions (e.g., constant, uniform, weibull, normal, etc.) - custom distributions (e.g., ones created using the PDF editor) - scripted distributions (draw outcome from values specified in a file) - provides control on what kind of outcomes are needed (e.g., oms_dist_nonnegative_outcome () will give only positive values from a distribution -- this is useful when the distribution parameters are such that the outcomes can be negative whereas a negative outcome may not mean anything for a parameters such as packet size, interarrival time etc.)3. The oms_dist package also provides hooks, sample code that show you how to create a custom distribution based on analytical equations. For example, check-out oms_dist_normal_box_mueller_outcome () defined in the oms_dist_support.ex.c external code file.

DES Kernel->Process Modeling/Coding