Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study
Recently, the Logic-in-Memory (LiM) concept has been widely studied in the literature. This paradigm represents one of the most efficient ways to solve the limitations of a Von Neumann’s architecture: by placing simple logic circuits inside or near a memory element, it is possible to obtai...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-02-01
|
| Series: | Journal of Low Power Electronics and Applications |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9268/10/1/7 |
| _version_ | 1828273232442032128 |
|---|---|
| author | Andrea Coluccio Marco Vacca Giovanna Turvani |
| author_facet | Andrea Coluccio Marco Vacca Giovanna Turvani |
| author_sort | Andrea Coluccio |
| collection | DOAJ |
| description | Recently, the Logic-in-Memory (LiM) concept has been widely studied in the literature. This paradigm represents one of the most efficient ways to solve the limitations of a Von Neumann’s architecture: by placing simple logic circuits inside or near a memory element, it is possible to obtain a local computation without the need to fetch data from the main memory. Although this concept introduces a lot of advantages from a theoretical point of view, its implementation could introduce an increasing complexity overhead of the memory itself, leading to a more sophisticated design flow. As a case study, Binary Neural Networks (BNNs) have been chosen. BNNs binarize both weights and inputs, transforming multiply-and-accumulate into a simpler bitwise logical operation while maintaining high accuracy, making them well-suited for a LiM implementation. In this paper, we present two circuits implementing a BNN model in CMOS technology. The first one, called Out-Of-Memory (OOM) architecture, is implemented following a standard Von Neumann structure. The same architecture was redesigned to adapt the critical part of the algorithm for a modified memory, which is also capable of executing logic calculations. By comparing both OOM and LiM architectures we aim to evaluate if Logic-in-Memory paradigm is worth it. The results highlight that LiM architectures have a clear advantage over Von Neumann architectures, allowing a reduction in energy consumption while increasing the overall speed of the circuit. |
| first_indexed | 2024-04-13T06:17:56Z |
| format | Article |
| id | doaj.art-38b3155f9f6a47ce9a8e9c0aa60845db |
| institution | Directory Open Access Journal |
| issn | 2079-9268 |
| language | English |
| last_indexed | 2024-04-13T06:17:56Z |
| publishDate | 2020-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Low Power Electronics and Applications |
| spelling | doaj.art-38b3155f9f6a47ce9a8e9c0aa60845db2022-12-22T02:58:45ZengMDPI AGJournal of Low Power Electronics and Applications2079-92682020-02-01101710.3390/jlpea10010007jlpea10010007Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case StudyAndrea Coluccio0Marco Vacca1Giovanna Turvani2Department of electronics and telecommunication (DET), Politecnico di Torino, Corso Castelfidardo 39, 10129 Torino, ItalyDepartment of electronics and telecommunication (DET), Politecnico di Torino, Corso Castelfidardo 39, 10129 Torino, ItalyDepartment of electronics and telecommunication (DET), Politecnico di Torino, Corso Castelfidardo 39, 10129 Torino, ItalyRecently, the Logic-in-Memory (LiM) concept has been widely studied in the literature. This paradigm represents one of the most efficient ways to solve the limitations of a Von Neumann’s architecture: by placing simple logic circuits inside or near a memory element, it is possible to obtain a local computation without the need to fetch data from the main memory. Although this concept introduces a lot of advantages from a theoretical point of view, its implementation could introduce an increasing complexity overhead of the memory itself, leading to a more sophisticated design flow. As a case study, Binary Neural Networks (BNNs) have been chosen. BNNs binarize both weights and inputs, transforming multiply-and-accumulate into a simpler bitwise logical operation while maintaining high accuracy, making them well-suited for a LiM implementation. In this paper, we present two circuits implementing a BNN model in CMOS technology. The first one, called Out-Of-Memory (OOM) architecture, is implemented following a standard Von Neumann structure. The same architecture was redesigned to adapt the critical part of the algorithm for a modified memory, which is also capable of executing logic calculations. By comparing both OOM and LiM architectures we aim to evaluate if Logic-in-Memory paradigm is worth it. The results highlight that LiM architectures have a clear advantage over Von Neumann architectures, allowing a reduction in energy consumption while increasing the overall speed of the circuit.https://www.mdpi.com/2079-9268/10/1/7logic-in-memory (lim)von neumann’s bottleneckmemory-wall |
| spellingShingle | Andrea Coluccio Marco Vacca Giovanna Turvani Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study Journal of Low Power Electronics and Applications logic-in-memory (lim) von neumann’s bottleneck memory-wall |
| title | Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study |
| title_full | Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study |
| title_fullStr | Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study |
| title_full_unstemmed | Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study |
| title_short | Logic-in-Memory Computation: Is It Worth it? A Binary Neural Network Case Study |
| title_sort | logic in memory computation is it worth it a binary neural network case study |
| topic | logic-in-memory (lim) von neumann’s bottleneck memory-wall |
| url | https://www.mdpi.com/2079-9268/10/1/7 |
| work_keys_str_mv | AT andreacoluccio logicinmemorycomputationisitworthitabinaryneuralnetworkcasestudy AT marcovacca logicinmemorycomputationisitworthitabinaryneuralnetworkcasestudy AT giovannaturvani logicinmemorycomputationisitworthitabinaryneuralnetworkcasestudy |