Seongmoo Heo

Reducing power density through activity migration

Power dissipation is unevenly distributed in modern microprocessors leading to localized hot spots with significantly greater die temperature than surrounding cooler regions. Excessive junction temperature reduces reliability and can lead to catastrophic failure. We examine the use of activity migration which reduces peak junction temperature by moving computation between multiple replicated units. Using a thermal model that includes the temperature dependence of leakage power, we show that sustainable power dissipation can be increased by nearly a factor of two for a given junction temperature limit. Alternatively, peak die temperature can be reduced by 12.4°C at the same clock frequency. The model predicts that migration intervals of around 20--200 are required to achieve the maximum sustainable power increase. We evaluate several different forms of replication and migration policy control.

Dynamic fine-grain leakage reduction using leakage-biased bitlines

Leakage power is dominated by critical paths, and hence dynamic deactivation of fast transistors can yield large savings. We introduce metrics for comparing fine-grain dynamic deactivation techniques that include the effects of deactivation energy and startup latencies, as well as long-term leakage current. We present a new circuit-level technique for leakage current reduction, leakage-biased bitlines, that has low deactivation energy and fast wakeup times. We show how this technique can be applied at a fine grain within an active microprocessor, and how microarchitectural scheduling policies can improve its performance. Using leakage-biased bitlines to deactivate SRAM read paths within I-cache memories saves over 24% of leakage energy and 22% of total I-cache energy when using a 70nm process. In the register file, fine-grained read port deactivation saves nearly 50% of leakage energy and 22% of total energy. Independently, turning off idle register file subbanks saves over 67% of leakage energy (57% total register file energy) with no loss in performance.

Activity-Sensitive Flip-Flop and Latch Selection for Reduced Energy

This paper presents new techniques to evaluate the energy and delay of flip-flop and latch designs and shows that no single existing design performs well across the wide range of operating regimes present in complex systems. We propose the use of a selection of flip-flop and latch designs, each tuned for different activation patterns and speed requirements. We illustrate our technique on a pipelined MIPS processor datapath running SPECint95 benchmarks, where we reduce total flip-flop and latch energy by over 60% without increasing cycle time.

Leakage-biased domino circuits for dynamic fine-grain leakage reduction

A leakage-biased domino circuit family is proposed that maintains high speed in active mode but which can be rapidly placed into a low-leakage inactive state by using leakage currents themselves to bias internal nodes. A 32-bit Han-Carlson domino adder circuit is used to compare LB-domino with conventional single and dual Vt domino circuits. For equal delay and noise margin, the LB-domino technique gives two decades reduction in steady-state leakage energy compared to a dual-Vt technique.

Load-sensitive flip-flop characterizations

Different flip-flop designs vary in the number and complexity of logic stages they contain, and hence have different inherent parasitic delays and output drive strengths. We examine the effect of electrical load on flip-flop delay and energy consumption and show that the relative ranking of optimized flip-flop structures varies widely with both electrical effort and absolute load. We also show that some structures benefit substantially from the addition of appropriate output buffering.

Activity-sensitive flip-flop and latch selection for reduced energy

This paper presents new techniques to evaluate the energy and delay of flip-flop and latch designs and shows that no single existing design performs well across the wide range of operating regimes present in complex systems. We propose the use of a selection of flip-flop and latch designs, each tuned for different activation patterns and speed requirements. We illustrate our technique on a pipelined MIPS processor datapath running SPECint95 benchmarks, where we reduce total flip-flop and latch energy by over 60% without increasing cycle time.This article presents new techniques to evaluate the energy and delay of flip-flop and latch designs and shows that no single existing design performs well across the wide range of operating regimes present in complex systems. We prepose the use of a selection of flip-flop latch designs, each timed for different activation patterns and speed requirements. We illustrate the use of our technique on a pipelined MIPS processor datapath running SPECint95 benchmarks, where we reduce total flip-flop and latch energy by 60% without increasing cycle time.