Fundamentals of modern VLSI devices
Fundamentals of modern VLSI devices
Low power and high performance design challenges in future technologies
GLSVLSI '00 Proceedings of the 10th Great Lakes symposium on VLSI
Gated-Vdd: a circuit technique to reduce leakage in deep-submicron cache memories
ISLPED '00 Proceedings of the 2000 international symposium on Low power electronics and design
Cache decay: exploiting generational behavior to reduce cache leakage power
ISCA '01 Proceedings of the 28th annual international symposium on Computer architecture
Drowsy caches: simple techniques for reducing leakage power
ISCA '02 Proceedings of the 29th annual international symposium on Computer architecture
On the Comparison of IDDQ and IDDQ Testing
VTS '99 Proceedings of the 1999 17TH IEEE VLSI Test Symposium
Leakage current reduction in CMOS VLSI circuits by input vector control
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
A process-tolerant cache architecture for improved yield in nanoscale technologies
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Statistical analysis of SRAM cell stability
Proceedings of the 43rd annual Design Automation Conference
An IC manufacturing yield model considering intra-die variations
Proceedings of the 43rd annual Design Automation Conference
Process variation aware cache leakage management
Proceedings of the 2006 international symposium on Low power electronics and design
Transactions on high-performance embedded architectures and compilers III
Hi-index | 0.00 |
We observe that the same SRAM cell leaks differently, under withindie process variations, when storing 0 and 1; this difference can be up to 3 orders of magnitude (averaging 57%) at 60mv variation of threshold voltage (Vth). Thus, leakage can be reduced if most often the values with less leakage are stored in the cache SRAM cells. We show applicability of this proposal by presenting three binary-optimization and software-level techniques for reducing instruction cache leakage: we (i) reorder instructions within basic-blocks so as to match up the instructions with the less-leaky state of their corresponding cache cells, (ii) statically apply register-renaming with the same aim, and (iii) at boot time, initialize unused cache-lines to their corresponding less-leaky values. Experimental results show up to 54%, averaging 37%, leakage energy reduction at 60mv variation in Vth, and show that with technology scaling, this saving can reach up to 84% at 100mv Vth variation. Since our techniques are one-off and do not affect instruction cache hit ratio, this reduction is provided with only a negligible penalty, in rare cases, in the data cache.