Litcius/Paper detail

Optimized Two-Step Store Control for MTJ-Based Nonvolatile Flip-Flops to Minimize Store Energy Under Process and Temperature Variations

Kimiyoshi Usami, Daiki Yokoyama, Aika Kamei, Hideharu Amano, Kenta Suzuki, Keizo Hiraga, Kazuhiro Bessho

2023IEEE Transactions on Very Large Scale Integration (VLSI) Systems10 citationsDOI

Abstract

Introducing a magnetic tunneling junction (MTJ) into a flip-flop enables nonvolatile power gating (PG) but large store energy to MTJ is a critical concern. We propose an optimized two-step store (TSS) control to first perform a short store with an optimal time for all nonvolatile flip-flops (NVFFs) and then perform a long store only at the failed ones for reducing the store energy. As the key technologies to realize this, we present a verify-and-retryable NVFF (VR-NVFF) circuit enabling the TSS control and an analytical expression for the optimal short-store time ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${T}_{\text {short}{\_}\text{opt}}$ </tex-math></inline-formula> ) minimizing the store energy. To examine the effectiveness of the optimized TSS control and the validity of analytically derived <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${T}_{\text {short}{\_}\text{opt}}$ </tex-math></inline-formula> , we implemented the TSS control on a coarse-grained reconfigurable array (CGRA)-based accelerator chip and fabricated it in a 40-nm CMOS/MTJ hybrid process technology. Results demonstrated that analytical <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${T}_{\text {short}{\_}\text{opt}}$ </tex-math></inline-formula> showed a good agreement with the measured value (within 8% difference) under process and temperature variations. The TSS control with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${T}_{\text {short}{\_}\text{opt}}$ </tex-math></inline-formula> reduced the store energy to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0.32\times $ </tex-math></inline-formula> of that of the conventional long-store-only technique. The break-even time (BET), which is the minimum power-gating time to get the gain in energy savings, was shortened to 0.51– <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0.7\times $ </tex-math></inline-formula> by the TSS control, achieving the BET of 50– <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$923 \mu \text{s}$ </tex-math></inline-formula> in the range of 0 °C–80 °C.

Topics & Concepts

Energy (signal processing)Computer scienceProcess (computing)NotationCMOSAlgorithmArithmeticComputer hardwareElectrical engineeringMathematicsEngineeringProgramming languageStatisticsAdvancements in Semiconductor Devices and Circuit DesignSemiconductor materials and devicesEnergy Harvesting in Wireless Networks