Partial evaluation and automatic program generation
Partial evaluation and automatic program generation
Performance specification using attributed grammars
DAC '93 Proceedings of the 30th international Design Automation Conference
Performance Modeling Using PDL
Computer
Performance Specification and Measurement
CHDL '93 Proceedings of the 11th IFIP WG10.2 International Conference sponsored by IFIP WG10.2 and in cooperation with IEEE COMPSOC on Computer Hardware Description Languages and their Applications
| Hi-index | 0.00 |
Often during the design process, it is necessary to analyzethe effects of tradeoff among various performance attributesof the design.Visual representations in the formof plots, graphs, and tables are typically used.These visualizationscan be generated using different applications suchas MatLab®, Mathematica®, Excel®, and so forth.Inorder to use these tools, performance equations for the designmust be rendered in an equational form with only afew variables, usually 2-3.However, performance modelsusually consist of a large number of attributes and evaluzationprocedures, typically written using the full power of aprogramming or hardware description language.We introducea symbolic evaluation procedure to simplify performancemodels.Given partial data, such evaluation yields a residualperformance model which is simpler than the original model.Symbolic evaluation can be effectively used to obtain residualperformance equations in terms of the variables whosetradeoff visualization is of interest to the user.