ABSTRACT
This paper presents for the first time low energy simultaneousmemory and register allocation. A minimum cost network flowapproach is used to efficiently solve for minimum energy dissipationsolutions in polynomial time. Results show that estimatedenergy improvements of 1.4 to 2.5 times over previous researchare obtained. This research is important for industry since energydissipation is minimized without requiring an increase in cost.
- 1.K.Keutzer, "The Impact of CAD on the Design of Low Power Digital Circuits", IEEE Symposium on Low Power Electronics, 1994, p42-45.Google ScholarCross Ref
- 2.S.Wuytack,F.Catthoor,F.Franssen,L.Nachtergaele, H.DeMan, "Global Communication and Memory Optimizing Transformations For Low Power Systems", International Workshop on Low Power Design, 1994, p 203-208.Google Scholar
- 3.A.Chandrakasan,et.al."Optimizing Power Using Transformations", IEEE Transactions on CAD, Jan 1995,Vol14,No.1, p12- 31. Google ScholarDigital Library
- 4.A.Farrahi, G.Tellez,M.Sarrafzadeh, "Memory Segmentation to Exploit Sleep Mode Operation", Design Automation Conference, 1995. Google ScholarDigital Library
- 5.Garey and Johnson,Computers and Intractability New York Freeman and Co 1979. Google ScholarDigital Library
- 6.G.Chaitin, "Register Allocation & Spilling via Graph Coloring", ACM SIGPLAN Symp on Compiler Construction, 1982 Google ScholarDigital Library
- 7.F.Chow,J.Hennessey, "The Priority-Based Coloring Approach to Register Allocation", ACM Transactions on Programming Languages and Systems, p501-536, Oct 1990. Google ScholarDigital Library
- 8.J.Chang, M.Pedram, "Register Allocation and Binding for Low Power", Design Automation Conference, 1995. Google ScholarDigital Library
- 9.V.Tiwari, S.Malik,A.Wolfe, "Power Analysis of Embedded Software ;A First Step Towards Software Power Minimization, IEEE Trans on VLSI, Vol. 2, No. 4, Dec 1994, p437-445, Google ScholarDigital Library
- 10.D.Kolson,A.Nicolau,N.Dutt,K.Kennedy, "Optimal Register Allocation to Loops for Embedded Code Generation", International Symposium on Systems Synthesis, p42-47, 1995. Google ScholarDigital Library
- 11.S.Wuytack,F.Catthoor,L.Nachtergaele,H.DeMan, "Power Exploration for Data Dominated Video Applications", ISLPED, p359-364, 1996. Google ScholarDigital Library
- 12.W.Cheng, Y-L.Lin, "A Transformation-Based Approach for Storage Optimization", ACM/IEEE Design Automation Conference, 1995. Google ScholarDigital Library
- 13.P.Landman, J.Rabaey, "Activity-Sensitive Architectural Power Analysis", IEEE Transactions on CAD,Vol.15,No.6, June 1996. Google ScholarDigital Library
- 14.P.A.Beerel, USC, Panel Presentation on "Where Does the Power GOT" at International Symposium on Low Power Design, April 1995.Google Scholar
- 15.Lee,Tiwari, "A Memory Allocation Technique for Low- Energy Embedded DSP Software", Symposium on Low Power Electronics, Oct, 1995, p24-5.Google Scholar
- 16.Lidsky,Rabaey, "Low-Power Design of Memory Intensive Functions", IEEE Symposium on Low Power Electronics, 1994, p16-7.Google Scholar
- 17.Nemhauser, Wolsey, Integer and Combinatorial Optimization, New York Wiley Interscience, 1988. Google ScholarDigital Library
- 18.H.Schmitt,D.Thomas, "Array Mapping in Behavioral Synthesis ", Int'l Symp on Systems Synthesis, 1995. Google ScholarDigital Library
- 19.P.Panda, N.Dutt, "Low Power Mapping of Behavioral Arrays to Multiple Memories", Int'l Symp on Low Power Electronic Design, p289-292, 1996. Google ScholarDigital Library
- 20.C.Gebotys, "Low Energy Memory Component Design for Cost-Sensitive High Performance Embedded Systems", IEEE Custom Integrated Circuits Conference, p397-400, 1996.Google ScholarCross Ref
- 21.C.Gebotys,R.Gebotys, "Performance-Power Optimization of Memory Components for Complex Embedded Systems", IEEE, 30th Hawaii International Conference on System Sciences, ECCS-6, Jan. 1997. Google ScholarDigital Library
Index Terms
- Low energy memory and register allocation using network flow
Recommendations
Efficient global register allocation for minimizing energy consumption
Data referencing during program execution can be a significant source of energy consumption especially for data-intensive programs. In this paper, we propose an approach to minimize such energy consumption by allocating data to proper registers and ...
Bytewise Register Allocation
SCOPES '15: Proceedings of the 18th International Workshop on Software and Compilers for Embedded SystemsTraditionally, variables have been considered as atoms by register allocation: Each variable was to be placed in one register, or spilt (placed in main memory) or rematerialized (recalculated as needed). Some flexibility arose from what would be ...
Differential register allocation
PLDI '05: Proceedings of the 2005 ACM SIGPLAN conference on Programming language design and implementationMicro-architecture designers are very cautious about expanding the number of architected registers (also the register field), because increasing the register field adds to the code size, raises I-cache and memory pressure, complicates processor ...
Comments