On the Optimal Refresh Power Allocation for Energy-Efficient Memories

@article{Kim2019OnTO,
  title={On the Optimal Refresh Power Allocation for Energy-Efficient Memories},
  author={Yongjune Kim and Won Ho Choi and Cyril Guyot and Yuval Cassuto},
  journal={2019 IEEE Global Communications Conference (GLOBECOM)},
  year={2019},
  pages={1-6}
}
Refresh is an important operation to prevent loss of data in dynamic random-access memory (DRAM). However, frequent refresh operations incur considerable power consumption and degrade system performance. Refresh power cost is especially significant in high-capacity memory devices and battery-powered edge/mobile applications. In this paper, we propose a principled approach to optimizing the refresh power allocation. Given a model for the bit error rate dependence on power, we formulate a convex… 

Figures and Tables from this paper

Optimizing the Write Fidelity of MRAMs

This paper considers the mean squared error (MSE) as a fidelity metric and proposes an iterative water-filling algorithm to minimize the MSE, and formulate an optimization problem to maximize the memory fidelity given energy constraints.

Optimizing Write Fidelity of MRAMs by Alternating Water-Filling Algorithm

A biconvex optimization problem to optimize write fidelity given energy and latency constraints is formulated and it is proved that the algorithm can reduce the MSE exponentially with the number of bits per word.

Optimizing Write Fidelity of MRAMs via Iterative Water-filling Algorithm

The numerical results show that the optimized write pulses can achieve 40% write energy reduction for a given classification accuracy, and it is proved that the proposed algorithm can reduce the MSE exponentially with the number of bits per word.

References

SHOWING 1-10 OF 27 REFERENCES

Block-based multiperiod dynamic memory design for low data-retention power

This paper investigates a novel scheme that relies on small refresh blocks and multiple refresh periods to reduce DRAM dissipation by decreasing the number of cells refreshed too often and gives a novel polynomial-time algorithm for computing an optimal set of refresh periods for block-based multiperiod refresh.

DRAM Refresh Mechanisms, Penalties, and Trade-Offs

This paper describes the difference in refresh operations between modern synchronous DRAM and traditional asynchronous DRAM; the refresh modes and timings; and variations in data retention time; and quantifies refresh penalties versus device speed, size, and total memory capacity.

Reducing cache power with low-cost, multi-bit error-correcting codes

The significant impact of variations on refresh time and cache power consumption for large eDRAM caches is shown and Hi-ECC, a technique that incorporates multi-bit error-correcting codes to significantly reduce refresh rate, is proposed.

Reducing Refresh Power in Mobile Devices with Morphable ECC

This paper proposes Morphable ECC, which reduces refresh operations during idle mode by 16x, memory power in Idle mode by 2X, while retaining performance within 2% of a system that does not use any ECC.

Flikker: saving DRAM refresh-power through critical data partitioning

Flikker exposes and leverages an interesting trade-off between energy consumption and hardware correctness, and shows that many applications are naturally tolerant to errors in the non-critical data, and in the vast majority of cases, the errors have little or no impact on the application's final outcome.

Generalized Water-Filling for Source-Aware Energy-Efficient SRAMs

This work proposes a principled approach to determine optimal non-uniform bit-line swings by formulating convex optimization problems and shows that energy-optimal swing assignment reduces energy consumption by half at a peak signal-to-noise ratio of 30 dB for an 8-bit accessed word.

Fault-tolerant refresh power reduction of DRAMs for quasi-nonvolatile data retention

An adaptive refresh rate controller was developed in order to ensure robustness against the variations in data retention time due to perturbation effects such as DRAM part-to-part variations, environmental changes and data pattern sensitivity, while at the same time minimizing power usage.

Smart Refresh: An Enhanced Memory Controller Design for Reducing Energy in Conventional and 3D Die-Stacked DRAMs

  • Mrinmoy GhoshH. Lee
  • Computer Science
    40th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 2007)
  • 2007
The basic concept behind the scheme is that a DRAM row that was recently read or written to by the processor does not need to be refreshed again by the periodic refresh operation, thereby eliminating excessive refreshes and the energy dissipated.

RAIDR: Retention-aware intelligent DRAM refresh

This paper proposes RAIDR (Retention-Aware Intelligent DRAM Refresh), a low-cost mechanism that can identify and skip unnecessary refreshes using knowledge of cell retention times and group DRAM rows into retention time bins and apply a different refresh rate to each bin.

Relaxing non-volatility for fast and energy-efficient STT-RAM caches

It is found that a pure STT-RAM cache hierarchy provides the best energy efficiency, though a hybrid design of SRAM-based L1 caches with reduced-retention STt-RAM L2 and L3 caches eliminates performance loss while still reducing the energy-delay product by more than 70%.