Optimal selection of supply voltages and level conversions during data path scheduling under resource constraints

In this paper we will consider how to select an optimal set of supply voltages and account for level conversion costs when optimizing the schedule of a resource dominated data path for minimum energy dissipation. An integer linear program (ILP) is presented for minimum energy schedules under latency, supply voltage, and resource constraints. The supply voltage assignment for each resource is modeled as fixed for all time. Schedules were generated for a variety of data path structures, resource and latency constraints. Resource constraints tended to limit the use of reduced supply voltages. With latency constraints loosened to 1.5/spl times/ minimum latency, unlimited resources, and two power supplies, energy savings ranged from 53% to 70% compared to 5 V operation. When resource constraints were applied, savings dropped to a range of 46% to 58%. Loosened latency constraints resulted in increased use of lower supply voltages. With resource constraints un-changed and latency constraints of 2/spl times/ minimum latency, energy savings increased to a range of 64% to 75%. In no case did three supplies decrease energy by more than 5% compared to two supplies.

[1]  Miodrag Potkonjak,et al.  Optimizing power using transformations , 1995, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[2]  S. Katkoori,et al.  Profile-driven behavioral synthesis for low-power VLSI systems , 1995, IEEE Design & Test of Computers.

[3]  J. E. Mitchell,et al.  Analyzing and exploiting the structure of the constraints in the ILP approach to the scheduling problem , 1994, IEEE Trans. Very Large Scale Integr. Syst..

[4]  Niraj K. Jha,et al.  An iterative improvement algorithm for low power data path synthesis , 1995, Proceedings of IEEE International Conference on Computer Aided Design (ICCAD).

[5]  Catherine H. Gebotys,et al.  An optimal methodology for synthesis of DSP multichip architectures , 1995, J. VLSI Signal Process..

[6]  John G. Proakis,et al.  Digital Signal Processing: Principles, Algorithms, and Applications , 1992 .

[7]  Anantha P. Chandrakasan,et al.  Low-power CMOS digital design , 1992 .

[8]  Niraj K. Jha,et al.  Behavioral synthesis for low power , 1994, Proceedings 1994 IEEE International Conference on Computer Design: VLSI in Computers and Processors.

[9]  Catherine H. Gebotys,et al.  Optimal VLSI Architectural Synthesis , 1992 .

[10]  Mohamed I. Elmasry,et al.  Global optimization approach for architectural synthesis , 1990, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[11]  Anantha P. Chandrakasan,et al.  Design of portable systems , 1994, Proceedings of IEEE Custom Integrated Circuits Conference - CICC '94.

[12]  Alex Orailoglu,et al.  Microarchitectural synthesis of performance-constrained, low-power VLSI designs , 1994, Proceedings 1994 IEEE International Conference on Computer Design: VLSI in Computers and Processors.

[13]  John P. Knight,et al.  Power-Profiler: Optimizing ASICs Power Consumption at the Behavioral Level , 1995, 32nd Design Automation Conference.

[14]  Majid Sarrafzadeh,et al.  Variable voltage scheduling , 1995, ISLPED '95.

[15]  Mark Horowitz,et al.  Clustered voltage scaling technique for low-power design , 1995, ISLPED '95.

[16]  Giovanni De Micheli,et al.  Synthesis and Optimization of Digital Circuits , 1994 .

[17]  Catherine H. Gebotys,et al.  Optimal VLSI Architectural Synthesis: Area, Performance and Testability , 1991 .