A1 Refereed original research article in a scientific journal
Hierarchical Agent Monitoring Design Approach towards Self-Aware Parallel Systems-on-Chip
Authors: Guang L, Nigussie E, Rantala P, Isoaho J, Tenhunen H
Publisher: ASSOC COMPUTING MACHINERY
Publication year: 2010
Journal: ACM Transactions in Embedded Computing Systems
Journal name in source: ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS
Journal acronym: ACM T EMBED COMPUT S
Article number: ARTN 25
Number in series: 3
Volume: 9
Issue: 3
Number of pages: 24
ISSN: 1539-9087
DOI: https://doi.org/10.1145/1698772.1698783(external)
Abstract
Hierarchical agent framework is proposed to construct a monitoring layer towards self-aware parallel systems-on-chip (SoCs). With monitoring services as a new design dimension, systems are capable of observing and reconfiguring themselves dynamically at all levels of granularity, based on application requirements and platform conditions. Agents with hierarchical priorities work adaptively and cooperatively to maintain and improve system performance in the presence of variations and faults. Function partitioning of agents and hierarchical monitoring operations on parallel SoCs are analyzed. Applying the design approach on the Network-on-Chip (NoC) platform demonstrates the design process and benefits using the novel approach.
Hierarchical agent framework is proposed to construct a monitoring layer towards self-aware parallel systems-on-chip (SoCs). With monitoring services as a new design dimension, systems are capable of observing and reconfiguring themselves dynamically at all levels of granularity, based on application requirements and platform conditions. Agents with hierarchical priorities work adaptively and cooperatively to maintain and improve system performance in the presence of variations and faults. Function partitioning of agents and hierarchical monitoring operations on parallel SoCs are analyzed. Applying the design approach on the Network-on-Chip (NoC) platform demonstrates the design process and benefits using the novel approach.