Power and Energy Impact by Loop Transformations

In this paper, we study the contribution of compiler optimizations to energy reduction. In particular, we are interested in the impact of loop optimizations in terms of performance and power tradeoffs. Both low-level loop optimizations at code generation (back-end) phase, such as loop unrolling and software pipelining, and high-level loop optimizations at program analysis and transformation phase ( frontend), such as loop permutation and tiling, are studied.

[1]  W. Robert Daasch,et al.  TEM2P2EST: A Thermal Enabled Multi-model Power/Performance ESTimator , 2000, PACS.

[2]  Barbara M. Chapman,et al.  Supercompilers for parallel and vector computers , 1990, ACM Press frontier series.

[3]  Srinivas Mantripragada,et al.  A new framework for integrated global local scheduling , 1998, Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192).

[4]  Mahmut T. Kandemir,et al.  Influence of compiler optimizations on system power , 2001, IEEE Trans. Very Large Scale Integr. Syst..

[5]  Vivek Sarkar.  Optimized unrolling of nested loops , 2000, ICS '00.

[6]  B. Ramakrishna Rau,et al.  Some scheduling techniques and an easily schedulable horizontal architecture for high performance scientific computing , 1981, MICRO 14.

[7]  Todd M. Austin,et al.  The SimpleScalar tool set, version 2.0 , 1997, CARN.

[8]  Steven S. Muchnick,et al.  Advanced Compiler Design and Implementation , 1997 .

[9]  Raymond Lo,et al.  Register promotion by sparse partial redundancy elimination of loads and stores , 1998, PLDI.

[10]  John L. Hennessy,et al.  The priority-based coloring approach to register allocation , 1990, TOPL.

[11]  Michael Wolfe,et al.  Advanced Loop Interchanging , 1986, ICPP.

[12]  Raymond Lo,et al.  A new algorithm for partial redundancy elimination based on SSA form , 1997, PLDI '97.

[13]  Dirk Grunwald,et al.  A Comparison of Two Architectural Power Models , 2000, PACS.