Entropy-Based Low Power Data TLB Design

DMCA

Entropy-Based Low Power Data TLB Design (2006)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Chinnakrishnan Ballapuram , Gabriel H. Loh , Kiran Puttaswamy , Hsien-hsin S. Lee

Venue:	In Proc. of the Intnl. Conf. on Compilers, Architecture, and Synthesis for Embedded Systems
Citations:	2 - 1 self

BibTeX

@INPROCEEDINGS{Ballapuram06entropy-basedlow,
    author = {Chinnakrishnan Ballapuram and Gabriel H. Loh and Kiran Puttaswamy and Hsien-hsin S. Lee},
    title = {Entropy-Based Low Power Data TLB Design},
    booktitle = {In Proc. of the Intnl. Conf. on Compilers, Architecture, and Synthesis for Embedded Systems},
    year = {2006}
}

OpenURL

Abstract

The Translation Look-aside Buffer (TLB), a content addressable memory, consumes significant power due to the associative search mechanism it uses in the virtual to physical address translation. Based on our analysis of the TLB accesses, we make two observations. First, the entropy or information content of the stack virtual page numbers is low due to high spatial locality of stack memory references. Second, the entropy of the higher order bits of global memory references is low since the size of the global data is determined and fixed during compilation of a program. Based on these two characteristics, we propose two techniques: an entropy-based speculative stack address TLB and a deterministic global address TLB to achieve energy reducing. Our results show an average of 47 % energy savings in the data TLB with less than 1 % overall performance impact.

Keyphrases

associative search mechanism information content energy saving physical address translation order bit translation look-aside buffer global data data tlb high spatial locality overall performance impact stack virtual page number entropy-based speculative stack address tlb global memory reference significant power tlb access deterministic global address tlb content addressable memory stack memory reference

DMCA