Reducing Power Consumption of Embedded Dynamic Memories with ECCs

2026-07-17

AI Summary

This work addresses the high power consumption of embedded dynamic memory caused by frequent refresh operations necessitated by worst-case cells. To mitigate this issue, the authors propose an adaptive ECC selection scheme optimized for power efficiency. By establishing a refresh interval model and a power analysis framework, the method dynamically selects the optimal ECC configuration under a given yield constraint, effectively balancing the access overhead introduced by ECC against the energy savings from reduced refresh frequency. Experimental results demonstrate that, compared to a baseline without ECC, the proposed approach reduces total power consumption by 46.8% to 94.8% across various workloads, substantially improving energy efficiency.

Abstract

Gain-cell embedded dynamic random-access memory (GCRAM) offers dense and energy-efficient on-chip storage, but retention-time variations force frequent refresh operations to cover worst-case bits. Error-correction codes (ECCs) can alleviate this limitation by masking bit errors from weak cells and thereby reduce refresh cost. However, the trade-off between the additional access and logic energy introduced by ECCs and the power savings from longer refresh intervals is nontrivial, especially considering the wide range of available ECC options. To optimize overall power consumption, we propose an ECC selection method that combines a refresh-interval model with power analysis to identify the minimum-power ECC configurations under a given yield constraint. Across different memory bandwidths, activity factors, and read/write ratios, the evaluation results show that the best ECC option shifts from stronger codes in refresh-dominated operating regions to lower-overhead codes in access-dominated regions and achieves 46.8% to 94.8% reduction in total power relative to the no-ECC reference.

Problem

• embedded memory
• power consumption
• ECC
• refresh overhead
• retention time

Innovation

• ECC selection
• GCRAM
• refresh optimization
• power minimization
• embedded memory