Hardware Linking for FPGAs

Present design flows for digital hardware systems require many costly iterations for simulation, synthesis and the physical implementation on the chip. The productivity in designing such systems increases much slower than the progress in silicon industry, which is called the design productivity gap. Hence, the engineering cost of a product may overtake the production cost in the future. This holds especially for FPGAs (Field Programmable Gate Arrays), which are commonly used in lower volume products.

The aim of this master thesis is to develop and implement a component-based design methodology and a corresponding tool where systems are built with the help of predesigned components.  

 

Comparable to a linker that puzzles together multiple object files to a final program binary, completely implemented submodules will be integrated to a final system. When now changes occur to a submodule, this will then not influence other parts of the system. 

Hardware linking can be performed on different abstraction levels, including linking the configuration data or merging netlists that state a digital circuit. While manipulating configuration data can be easily performed at run-time, merging netlists will allow to simulate the linked system.
 

Theory (20%), Concepts (40%), Software implementation (40%) 

Prerequisites: INF1400, INF 2270, INF3430/4430, INF5430
Apportionment:  
Supervisors: Dirk Koch, Jim Tørresen


 

 

Tags: Hardware Linking for FPGAs
Published Aug. 1, 2011 4:12 AM - Last modified Oct. 24, 2011 9:33 PM