Optimizing Energy During Systems Synthesis of Computer Intensive Realtime Applications

Optimizing energy during the synthesis of VLSI systems for realtime-constrainedembedded applications is an important new problem. This paper presents a newmethodology for simultaneous scheduling and allocation of VLSI systems whichminimize estimated energy for large realtime compute intensive applications.Minimization of estimated energy and VLSI chip area using hierarchical decomposition,bin packing algorithms and integer linear programming techniques along with voltagescaling is performed. Common subexpression elimination, precomputation, dataregeneration, and loop merging transformations are supported. A large complex realindustrial application, audio compression, donated by Motorola, is used to study theenergy savings using different single and multichip system implementations. Results ofsynthesizing this complex application show that up to 10 times improvement inestimated energy are attainable for only 2.7 times increase in estimated chip area.Precomputation and other low energy transformations provided on average over 1.6times savings in energy respectively. This research is important for industry since energydissipation consideration at the early stages of design is crucial for mapping highperformance applications into cost-efficient and reliable systems