Memristive Linear Algebra
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911012756652032 |
|---|---|
| author | Lin, Jonathan Barrows, Frank Caravelli, Francesco |
| author_facet | Lin, Jonathan Barrows, Frank Caravelli, Francesco |
| contents | The advent of memristive devices offers a promising avenue for efficient and scalable analog computing, particularly for linear algebra operations essential in various scientific and engineering applications. This paper investigates the potential of memristive crossbars in implementing matrix inversion algorithms. We explore both static and dynamic approaches, emphasizing the advantages of analog and in-memory computing for matrix operations beyond multiplication. Our results demonstrate that memristive arrays can significantly reduce computational complexity and power consumption compared to traditional digital methods for certain matrix tasks. Furthermore, we address the challenges of device variability, precision, and scalability, providing insights into the practical implementation of these algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_20539 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Memristive Linear Algebra Lin, Jonathan Barrows, Frank Caravelli, Francesco Mesoscale and Nanoscale Physics Distributed, Parallel, and Cluster Computing Classical Analysis and ODEs Dynamical Systems Adaptation and Self-Organizing Systems The advent of memristive devices offers a promising avenue for efficient and scalable analog computing, particularly for linear algebra operations essential in various scientific and engineering applications. This paper investigates the potential of memristive crossbars in implementing matrix inversion algorithms. We explore both static and dynamic approaches, emphasizing the advantages of analog and in-memory computing for matrix operations beyond multiplication. Our results demonstrate that memristive arrays can significantly reduce computational complexity and power consumption compared to traditional digital methods for certain matrix tasks. Furthermore, we address the challenges of device variability, precision, and scalability, providing insights into the practical implementation of these algorithms. |
| title | Memristive Linear Algebra |
| topic | Mesoscale and Nanoscale Physics Distributed, Parallel, and Cluster Computing Classical Analysis and ODEs Dynamical Systems Adaptation and Self-Organizing Systems |
| url | https://arxiv.org/abs/2407.20539 |