BEARdocs

Two New Techniques for Unit-Delay Compiled Simulation

DSpace/Manakin Repository

Show simple item record

dc.contributor.author Maurer, Peter M.
dc.identifier.uri http://hdl.handle.net/2104/5449
dc.description.abstract The PC-set method and the parallel technique are two methods for generating compiled unit-delay simulations of acyclic circuits. The PC-set method analyzes the network, determines the set of potential change times for each net, and generates gate simulations for each potential change. The parallel technique, which is based on the concept of bit-parallel simulation, is faster and generates less code than the PC-set method, but is not amenable to data-parallel simulation of multiple input vectors. Both techniques are based on the well known levelization algorithm used to generate zero-delay Levelized Compiled Code simulations. Although the parallel technique provides for efficient simulations with a reasonable amount of generated code, there are several opportunities for optimization. The two optimization schemes presented here are called bit-field trimming and shift-elimination. Two different methods of shift elimination are presented, one which is based on breaking cycles in an undirected graph, and one which is based on tracing paths through a network. Performance results are presented for both simulation techniques, and for all optimizations. Comparisons with interpreted event-driven simulation using the ISCAS 85 benchmarks show a factor of 4 improvement for the PC-set method and a factor of ten improvement for the parallel technique. Similar studies for the optimization schemes show an average performance improvement of 47% over the unoptimized simulations. en
dc.subject Digital Simulation en
dc.subject Compiled Simulation en
dc.subject Unit-Delay Simulation en
dc.title Two New Techniques for Unit-Delay Compiled Simulation en
dc.license GPL en


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search BEARdocs


Advanced Search

Browse

My Account

Statistics