| 摘要 |
<p>A process and system are provided for optimizing the evaluation of signals in a simulation model. Optimization is achieved by generating for each signal in the simulation a customized method chain for computing only those attributes which are used by the signal. A method chain is generated for a particular signal by first determining which attributes are used by the signal. This determination is performed by traversing the design hierarchy of the simulation model. A list of used attributes is thus derived. Thereafter, the method chain is generated by first providing a plurality of method implementation code fragments, each code fragment being a set of computer code designed to perform a selected function. Once the code fragments are provided, the next step is to determine which functions, and hence, which code fragments, are needed to compute the used attributes. A list of needed code fragments is derived by traversing the list of used attributes and determining which functions need to be performed for each attribute. Once it is known which code fragments are needed, a method data structures is generated to call each of the needed fragments. These data structures are linked together to form the method chain for the signal. A method chain formed using this process will compute only those attributes which are actually used by the signal in the simulation. Since no extraneous attributes are computed, signal evaluation is optimized. <IMAGE></p> |
