Magnetic Tunnel Junctions for Timekeeping in Intermittent Computing Systems

2026-07-24

AI Summary

This work addresses the challenge of time-state loss during power failures in batteryless intermittent computing systems, which undermines sensing, scheduling, and coordination. Existing capacitor-based timekeeping methods suffer from strong trade-offs among measurement range, energy consumption, and area, along with aging-induced drift. To overcome these limitations, the authors propose a battery-free timekeeping mechanism leveraging engineered “failed” magnetic tunnel junction (MTJ) arrays, whose geometrically determined, predictable stochastic state decay enables time estimation. This approach decouples range from energy overhead, eliminates aging effects, and integrates device-trajectory-driven simulation with a stochastic retention-loss inference algorithm. Implemented in just 1.03 μJ and under 0.1 mm², the system achieves over 15 minutes of power-off time tracking with less than 10% error—offering a 9.2× extended range and 11× lower energy than prior art—and reduces scheduling errors by 16–52× over a one-year deployment.

Abstract

Batteryless intermittent systems run unattended for years, but power failures erase timekeeping state, corrupting sensing, scheduling, and coordination. State-of-the-art timekeepers infer elapsed time from capacitor discharge; however, the capacitor must be sized for the longest interval measured (so range, energy, and area grow together), and repeated charge-discharge cycling lowers capacitance over time, biasing every estimate further as the deployment ages. We present FLINT, a timekeeper that reads elapsed time from the stochastic retention loss of an array of "broken" Magnetic Tunnel Junctions (MTJs)---spintronic memory cells engineered to lose state predictably. Because the decay timescale is fixed by device geometry, the energy to read it is independent of the interval measured and does not drift with device age. We validate FLINT's array model against 21 fabricated MTJs, then evaluate the full timekeeper in real-device-trace-driven simulation, showing that it tracks over 15 minutes of off-time within 10% error while consuming only 1.03 $μJ$ and occupying under 0.1 $mm^2$---$9.2\times$ the range at $11\times$ lower energy than prior work. It extends to longer intervals at no added cost, and makes $16-52\times$ fewer scheduling errors than an aging capacitor clock over a one-year deployment.

Problem

• intermittent computing
• timekeeping
• power failure
• capacitor aging
• state retention

Innovation

• Magnetic Tunnel Junction
• Intermittent Computing
• Batteryless Systems
• Timekeeping
• Stochastic Retention