L0 Cluster Synthesis and Operation Shuffling

Clustered L0 buffers are an interesting alternative for reducing energy consumption in the instruction memory hierarchy of embedded VLIW processors. Currently, the synthesis of L0 clusters is performed as an hardware optimization, where the compiler generates a schedule and based on the given schedule L0 clusters are synthesized. The result of clustering is schedule dependent, which offers a design space for exploring the effects on clustering by scheduling. This paper presents a study indicating the potentials offered by shuffling operations within a VLIW instruction on L0 cluster synthesis. The simulation results indicate that potentially up to 75% of L0 buffer energy can be reduced by shuffling operations.

[1]  Gustavo de Veciana,et al.  Application-specific clustered VLIW datapaths: early exploration on a parameterized design space , 2002, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[2]  Geoffrey Brown,et al.  Lx: a technology platform for customizable VLIW embedded processing , 2000, ISCA '00.

[3]  Gustavo de Veciana,et al.  Design Challenges for New Application-Specific Processors , 2000, IEEE Des. Test Comput..

[4]  Vittorio Zaccaria,et al.  An instruction-level methodology for power estimation and optimization of embedded VLIW cores , 2002, Proceedings 2002 Design, Automation and Test in Europe Conference and Exhibition.

[5]  Henk Corporaal,et al.  A Low Energy Clustered Instruction Memory Hierarchy for Long Instruction Word Processors , 2002, PATMOS.

[6]  Ira Krepchin,et al.  Texas Instruments Inc. , 1963, Nature.

[7]  Scott A. Mahlke,et al.  Trimaran: An Infrastructure for Research in Instruction-Level Parallelism , 2004, LCPC.

[8]  John Arends,et al.  Instruction fetch energy reduction using loop caches for embedded applications with small tight loops , 1999, ISLPED '99.

[9]  Raminder Singh Bajwa,et al.  Instruction buffering to reduce power in processors for signal processing , 1997, IEEE Trans. Very Large Scale Integr. Syst..

[10]  L. Benini,et al.  A Power Modeling and Estimation Framework for VLIW-based Embedded Systems , 2001 .

[11]  Margaret Martonosi,et al.  Wattch: a framework for architectural-level power analysis and optimizations , 2000, Proceedings of 27th International Symposium on Computer Architecture (IEEE Cat. No.RS00201).

[12]  Henk Corporaal,et al.  Clustered L0 Buffer Organization for Low Energy Embedded Processors , 2002 .