Scaling Limits of Memristor-Based Routers for Asynchronous Neuromorphic Systems

• URL: http://arxiv.org/abs/2307.08116v2
• Date: Thu, 21 Dec 2023 01:21:24 GMT
• Title: Scaling Limits of Memristor-Based Routers for Asynchronous Neuromorphic Systems
• Authors: Junren Chen, Siyao Yang, Huaqiang Wu, Giacomo Indiveri, Melika Payvand
• Abstract summary: Multi-core neuromorphic systems typically use on-chip routers to transmit spikes among cores. A promising alternative is to exploit the features of memristive crossbar arrays and use them as programmable switch-matrices that route spikes. We study the challenges of memristive crossbar arrays, when used as routing channels to transmit spikes in asynchronous Spiking Neural Network (SNN) hardware.
• Score: 2.5264231114078353
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-core neuromorphic systems typically use on-chip routers to transmit spikes among cores. These routers require significant memory resources and consume a large part of the overall system's energy budget. A promising alternative approach to using standard CMOS and SRAM-based routers is to exploit the features of memristive crossbar arrays and use them as programmable switch-matrices that route spikes. However, the scaling of these crossbar arrays presents physical challenges, such as "IR drop" on the metal lines due to the parasitic resistance, and leakage current accumulation on multiple active memristors in their "off" state. While reliability challenges of this type have been extensively studied in synchronous systems for compute-in-memory matrix-vector multiplication (MVM) accelerators and storage class memory, little effort has been devoted so far to characterizing the scaling limits of memristor-based crossbar routers. Here, we study the challenges of memristive crossbar arrays, when used as routing channels to transmit spikes in asynchronous Spiking Neural Network (SNN) hardware. We validate our analytical findings with experimental results obtained from a 4K-ReRAM chip which demonstrates its functionality as a routing crossbar. We determine the functionality bounds on the routing due to the IR drop and leak problem, based on theoretical modeling, circuit simulations for a 22nm FDSOI technology, and experimental measurements. This work highlights the limitations of this approach and provides useful guidelines for engineering the memristor device properties in memristive crossbar routers for multi-core asynchronous neuromorphic systems.

Related papers

• Neuromorphic Wireless Split Computing with Multi-Level Spikes [69.73249913506042]
  In neuromorphic computing, spiking neural networks (SNNs) perform inference tasks, offering significant efficiency gains for workloads involving sequential data. Recent advances in hardware and software have demonstrated that embedding a few bits of payload in each spike exchanged between the spiking neurons can further enhance inference accuracy. This paper investigates a wireless neuromorphic split computing architecture employing multi-level SNNs.
  arXiv  Detail & Related papers  (2024-11-07T14:08:35Z)
• Quantum random access memory with transmon-controlled phonon routing [8.07618328909056]
  Quantum random access memory (QRAM) promises simultaneous data queries at multiple memory locations. We introduce a transmon-controlled phonon router and propose a QRAM implementation by connecting these routers in a tree-like architecture.
  arXiv  Detail & Related papers  (2024-11-01T16:30:21Z)
• Multi-level, Forming Free, Bulk Switching Trilayer RRAM for Neuromorphic Computing at the Edge [0.0]
  We develop a forming-free and bulk switching RRAM technology based on a trilayer metal-oxide stack. We develop a neuromorphic compute-in-memory platform based on trilayer bulk RRAM crossbars. Our work paves the way for neuromorphic computing at the edge under strict size, weight, and power constraints.
  arXiv  Detail & Related papers  (2023-10-20T22:37:46Z)
• CktGNN: Circuit Graph Neural Network for Electronic Design Automation [67.29634073660239]
  This paper presents a Circuit Graph Neural Network (CktGNN) that simultaneously automates the circuit topology generation and device sizing. We introduce Open Circuit Benchmark (OCB), an open-sourced dataset that contains $10$K distinct operational amplifiers. Our work paves the way toward a learning-based open-sourced design automation for analog circuits.
  arXiv  Detail & Related papers  (2023-08-31T02:20:25Z)
• Extensible circuit-QED architecture via amplitude- and frequency-variable microwaves [52.77024349608834]
  We introduce a circuit-QED architecture combining fixed-frequency qubits and microwave-driven couplers. Drive parameters appear as tunable knobs enabling selective two-qubit coupling and coherent-error suppression.
  arXiv  Detail & Related papers  (2022-04-17T22:49:56Z)
• Working Memory Connections for LSTM [51.742526187978726]
  We show that Working Memory Connections constantly improve the performance of LSTMs on a variety of tasks. Numerical results suggest that the cell state contains useful information that is worth including in the gate structure.
  arXiv  Detail & Related papers  (2021-08-31T18:01:30Z)
• Energy-Efficient Model Compression and Splitting for Collaborative Inference Over Time-Varying Channels [52.60092598312894]
  We propose a technique to reduce the total energy bill at the edge device by utilizing model compression and time-varying model split between the edge and remote nodes. Our proposed solution results in minimal energy consumption and $CO$ emission compared to the considered baselines.
  arXiv  Detail & Related papers  (2021-06-02T07:36:27Z)
• Analog Neural Computing with Super-resolution Memristor Crossbars [0.0]
  Memristor crossbar arrays are used in a wide range of in-memory and neuromorphic computing applications. This paper presents a technique to improve the resolution by building a super-resolution memristor crossbar with nodes having multiple memristors. The wider the range and number of conductance values, the higher the crossbar's resolution.
  arXiv  Detail & Related papers  (2021-05-10T18:52:44Z)
• Endurance-Aware Mapping of Spiking Neural Networks to Neuromorphic Hardware [4.234079120512533]
  Neuromorphic computing systems are embracing memristors to implement high density and low power synaptic storage as crossbar arrays in hardware. Long bitlines and wordlines in a memristive crossbar are a major source of parasitic voltage drops, which create current asymmetry. We propose eSpine, a technique to improve lifetime by incorporating the endurance variation within each crossbar in mapping machine learning workloads.
  arXiv  Detail & Related papers  (2021-03-09T20:43:28Z)
• Reliability-Performance Trade-offs in Neuromorphic Computing [0.30458514384586394]
  Neuromorphic architectures built with Non-Volatile Memory (NVM) can significantly improve the energy efficiency of machine learning tasks designed with Spiking Neural Networks (SNNs) We observe that the parasitic voltage drops create a significant asymmetry in programming speed and reliability of NVM cells in a crossbar. This asymmetry in neuromorphic architectures create reliability-performance trade-offs, which can be exploited efficiently using SNN mapping techniques.
  arXiv  Detail & Related papers  (2020-09-26T19:38:18Z)
• DeepSIC: Deep Soft Interference Cancellation for Multiuser MIMO Detection [98.43451011898212]
  In multiuser multiple-input multiple-output (MIMO) setups, where multiple symbols are simultaneously transmitted, accurate symbol detection is challenging. We propose a data-driven implementation of the iterative soft interference cancellation (SIC) algorithm which we refer to as DeepSIC. DeepSIC learns to carry out joint detection from a limited set of training samples without requiring the channel to be linear.
  arXiv  Detail & Related papers  (2020-02-08T18:31:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.