Saved in:
Bibliographic Details
Main Authors: Kirschner, Maximilian, Dudzik, Konstantin, Krusekamp, Ben, Becker, Jürgen
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.05321
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908851592232960
author Kirschner, Maximilian
Dudzik, Konstantin
Krusekamp, Ben
Becker, Jürgen
author_facet Kirschner, Maximilian
Dudzik, Konstantin
Krusekamp, Ben
Becker, Jürgen
contents Real-time systems, particularly those used in domains like automated driving, are increasingly adopting neural networks. From this trend arises the need for high-performance hardware exhibiting predictable timing behavior. While state-of-the-art real-time hardware often suffers from limited memory and compute resources, modern AI accelerators typically lack the crucial predictability due to memory interference. We present a new hardware architecture to bridge this gap between performance and predictability. The architecture features a multi-core vector processor with predictable cores, each equipped with local scratchpad memories. A central management core orchestrates access to shared external memory following a statically determined schedule. To evaluate the proposed hardware architecture, we analyze different variants of our parameterized design. We compare these variants to a baseline architecture consisting of a single-core vector processor with large vector registers. We find that configurations with a larger number of smaller cores achieve better performance due to increased effective memory bandwidth and higher clock frequencies. Crucially for real-time systems, execution time fluctuation remains very low, demonstrating the platform's time predictability.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05321
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MultiVic: A Time-Predictable RISC-V Multi-Core Processor Optimized for Neural Network Inference
Kirschner, Maximilian
Dudzik, Konstantin
Krusekamp, Ben
Becker, Jürgen
Hardware Architecture
Real-time systems, particularly those used in domains like automated driving, are increasingly adopting neural networks. From this trend arises the need for high-performance hardware exhibiting predictable timing behavior. While state-of-the-art real-time hardware often suffers from limited memory and compute resources, modern AI accelerators typically lack the crucial predictability due to memory interference. We present a new hardware architecture to bridge this gap between performance and predictability. The architecture features a multi-core vector processor with predictable cores, each equipped with local scratchpad memories. A central management core orchestrates access to shared external memory following a statically determined schedule. To evaluate the proposed hardware architecture, we analyze different variants of our parameterized design. We compare these variants to a baseline architecture consisting of a single-core vector processor with large vector registers. We find that configurations with a larger number of smaller cores achieve better performance due to increased effective memory bandwidth and higher clock frequencies. Crucially for real-time systems, execution time fluctuation remains very low, demonstrating the platform's time predictability.
title MultiVic: A Time-Predictable RISC-V Multi-Core Processor Optimized for Neural Network Inference
topic Hardware Architecture
url https://arxiv.org/abs/2511.05321