Talk

Automatic Parallelization of the Visual Data-Flow Language Cantata for Efficient Characterization of Analog Circuit Behavior

Michael Goedecke,  Sorin A. Huss,  Kai Morich  

 About the Authors

  Keywords

Visual Characterization   Macro-Data-Flow   Parallelization   Load Balancing  

 Abstract

This paper presents a new approach to the characterization of analog circuits based on the visual data-flow language Cantata which has been augmented with some application specific functions and control operators. For the reduction of execution time a new data-flow scheduler for a distributed environment is proposed. The scheduler takes information about usable workstations and their performance, program availability and the actual workloads into consideration to achieve a load balancing which results in shorter execution times. This process is fully automated.

 Document in PostScript (compressed, 69027 bytes)

Automatic Construction of User Interfaces from Constraint Multiset Grammars	Bending the Rules: Steps toward Semantically enriched Graphical Rewrite Rules	Index of Talks	Proceedings - 11th Intl. Symposium on Visual Languages