Compiler-Directed Code Restructuring for Reducing Data TLB energy

DMCA

Compiler-Directed Code Restructuring for Reducing Data TLB energy (2004)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by M. Kandemir , I. Kadayif , G. Chen

Venue:	In Proceedings International Conference on Hardware/Software Codesign and System Synthesis
Citations:	4 - 0 self

BibTeX

@INPROCEEDINGS{Kandemir04compiler-directedcode,
    author = {M. Kandemir and I. Kadayif and G. Chen},
    title = {Compiler-Directed Code Restructuring for Reducing Data TLB energy},
    booktitle = {In Proceedings International Conference on Hardware/Software Codesign and System Synthesis},
    year = {2004}
}

OpenURL

Abstract

Prior work on TLB power optimization considered circuit and architectural techniques. A recent software-based technique for data TLBs has considered the possibility of storing the frequently used virtual-to-physical address translations in a set of translation registers (TRs), and using them when necessary instead of going to the data TLB. This paper presents a compiler-based strategy for increasing the effectiveness of TRs. The idea is to restructure the application code in such a fashion that once a TR is loaded, its contents are reused as much as possible. Our experimental evaluation with six array-based benchmarks from the Spec2000 suite indicates that the proposed TR reuse strategy brings significant reductions in data TLB energy over an alternate strategy that employs TRs but does not restructure the code for TR reuse

Keyphrases

data tlb energy compiler-directed code restructuring recent software-based technique spec2000 suite significant reduction compiler-based strategy translation register alternate strategy architectural technique data tlb virtual-to-physical address translation data tlbs tlb power optimization experimental evaluation tr reuse strategy prior work application code tr reuse array-based benchmark

DMCA