2026· ITM Web of Conferences· 0 citations· 12 references
TL;DR
This paper summarizes the trade-offs between power consumption, area and performance in the current technologies, and discusses the likely future trends including machine learning based intelligent routing and 3D integration.
Abstract
With the growing demands of the performance of multi-core systems, conventional bus architectures have posed a challenge to in-chip communication due to its low-scaling nature and challenges of globally synchronizing clocks. The Network-on-Chip (NoC) is based on the principles of the modern communication network and it offers a basic solution to high-throughput and low-latency communications. This paper systematically reviews major NoC technologies to meet these demands. It first introduces the state of research currently in terms of topology, routing algorithms, and router microstructure. In the analysis module, critical drawbacks of traditional systems are revealed, such as unbalanced load in mesh structures, congestion issues with deterministic routing, and inefficient resource use in common router structures. Finally, this paper summarizes the trade-offs between power consumption, area and performance in the current technologies, and discusses the likely future trends including machine learning based intelligent routing and 3D integration. This paper aims to offer a technical reference systematically on the designs of highperformance NoC in the future.
The scalability of multiprocessor networks on-chip is restricted by the performance of on-chip interconnects. Optical networks-on-chip (ONoCs) offer ultra-high bandwidth and energy-efficient solutions for these bottlenecks. Although 2D ONoCs have several advantages, scalability is a challenge because of limited chip ar...
Nathrao B. Jadhav, Bharat S. Chaudhari, P. Khandekar· Future Internet· 0 citations
A review of the DSP hardware architecture for URLLC scenarios and the collaboration of AI with DSP, terahertz processing, and 6G extreme URLLC architecture is envisioned, providing systematic technical references for researchers.
Hao-Zhe Huang· ITM Web of Conferences· 0 citations
Protecting the Network-on-Chip (NoC) of physical-AI tile-based accelerators deployed in harsh environments against single-event effects (SEEs) is paramount for preventing NoC failures that can lead to deadlocks and silent data corruption (SDC). Prior work on reliable NoCs has largely focused on narrow links (e.g., 32-b...
Chen Wu, M. Rogenmoser, Luca Benini et al.· 0 citations
Currently, as System-on-Chip (SoC) archiStectures scale in complexity, it is necessary to have an efficient communication between the components and also the demand for highthroughput, latency-tolerant memory interconnects has become critical. This paper presents the design, implementation and comprehensive functional...
Shail Nimeshkumar Shah, Sachin Gajjar· International Symposium on V...· 0 citations
This work compares two recent multicore neuromorphic systems implemented in the same 22-nm FDSOI technology and explicitly optimized for inter-core event communication, and discusses routing-aware training as a means of jointly optimizing neural connectivity, task performance, and hardware mappability.
Sebastian Billaudelle, Christian Metzner, Jimmy Weber et al.· 1 citation
As the scale of many-core systems-on-chip (SoC) continues to grow, application mapping in Network-on-Chip (NoC) architectures plays a crucial role in determining system-level communication efficiency and overall performance. However, existing mapping approaches often suffer from limited scalability, slow convergence, a...
Zhi Cheng, Yuan-Hao Zhao, Li-Xin He et al.· Journal of Circuits, Systems...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.