Pinning fault mode modeling for DWM shifting

 Roxy, K., Longofono, S., Olliver, S., Bhanja, S., & Jones, A. K. (2022). Pinning fault mode modeling for DWM shifting. IEEE Transactions on Circuits and Systems II: Express Briefs, 69(7), 3319-3323.

Abstract:

Extreme scaling for purposes of achieving higher density and lower energy continues to increase the probability of memory faults. For domain wall (DW) memories, misalignment faults arise when aligning domains with access points. A previously understudied type of shifting fault, a pinning fault may occur due to non-uniform pinning potential distribution caused by notches with fabrication imperfections. This non-uniformity can pin a wall during current-induced DW motion. This brief provides a model of geometric variations varying width, depth, and curvature variations of a notch, their impacts on the critical shift current, and a study of the resulting impact on fault rates of DW memory systems. An increase in the effective critical shift current due to 5% variation predicts a pinning fault rate on the order of 10 ^-8 per shift. This results in a mean-time-to-failure (MTTF) of circa 2s for a DW memory system and requires multi-bit error correction for achieving reasonable system lifetimes.

 

XDWM: A 2D Domain Wall Memory

 Hoque, A., Jones, A. K., & Bhanja, S. (2022). XDWM: A 2D Domain Wall Memory. IEEE Transactions on Nanotechnology, 21, 185-188.

 

Abstract:

Domain-Wall Memory (DWM) structures typically bundle nanowires shifted together for parallel access. Ironically, this organization does not allow the natural shifting of DWM to realize logical shifting within data elements. We describe a novel 2-D DWM cross-point (X-Cell) that allows two individual nanowires placed orthogonally to share the X-Cell. Each nanowire can operate independently while sharing the value at the X-Cell. Using X-Cells, we propose an orthogonal nanowire in the Y dimension overlaid on a bundle of X dimension nanowires for a cross-DWM or XDWM. We demonstrate that the bundle shifts correctly in the X-Direction, and that data can be logically shifted in the Y-direction providing novel data movement and supporting processing-in-memory. We conducted studies on the requirements for physical cell dimensions and shift currents for XDWM. Due to the non-standard domain, our micro-magnetic studies demonstrate that XDWM introduces a shift current penalty of 6.25% while shifting happens in one nanowire compared to a standard nanowire. We also demonstrate correct shifting using nanowire bundles in both the X- and Y- dimensions. Using magnetic simulation to derive the values for SPICE simulation we show the maximum leakage current between nanowires when shifting the bundle together is ≤ 3% indicating that sneak paths are not problematic for XDWM.

 

@ARTICLE{9735323,
  author={Hoque, Arifa and Jones, Alex K. and Bhanja, Sanjukta},
  journal={IEEE Transactions on Nanotechnology}, 
  title={XDWM: A 2D Domain Wall Memory}, 
  year={2022},
  volume={21},
  number={},
  pages={185-188},
  doi={10.1109/TNANO.2022.3158889}}

A Study on Reconfigurable Nanomagnetic Array and Effect of Gilbert Damping on Reconfigurability

 Hoque, A., Rajaram, S., & Bhanja, S. (2021). A Study on Reconfigurable Nanomagnetic Array and Effect of Gilbert Damping on Reconfigurability. IEEE Transactions on Nanotechnology, 20, 503-506.

 

Abstract:

A 2-D grid of circular nanomagnets (NMs) has been effectively used to solve binary quadratic optimization problems via magnetic interaction in the X-Y plane. Reconfiguring this architecture is challenging due to the geometry and spacing constraints imposed by the computational algorithm, but can be achieved by destabilizing the uniform and unidirectional magnetic influence from the non-computing cell. In this paper, we studied how the straightforward spin transfer torque (STT) can induce a rotational coupling field between the non-computing and computing cells to impair static interaction. This mechanism is also better than spin orbital torque (SOT)-based reconfiguration in terms of power, speed and system integration. Our study shows that STT-based solution requires 14× less current desnsity and is 25% faster than the SOT-based counterpart. We extended the study to achieve low-power reconfiguration by altering the Gilbert damping. A lower damping can further reduce 60% of the required current than the high damping case.

 

@ARTICLE{9448411,
  author={Hoque, Arifa and Rajaram, Srinath and Bhanja, Sanjukta},
  journal={IEEE Transactions on Nanotechnology}, 
  title={A Study on Reconfigurable Nanomagnetic Array and Effect of Gilbert Damping on Reconfigurability}, 
  year={2021},
  volume={20},
  number={},
  pages={503-506},
  doi={10.1109/TNANO.2021.3087590}}