Enhancing ARM-based Compute Core Performance using Task Priority and Proximity-Induced Thermal Thresholds in Heterogeneous Core Architecture
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Modern microprocessors featuring multi-core, multi-cluster CPU architectures incorporate multiple digital thermal sensors in heterogeneous cores, positioned within micrometer-scale proximity. Under high computational load, localized temperature elevation in an active core induces significant thermal coupling effects in adjacent cores, necessitating proactive thermal management. Experimental evaluation reveals that ARM-architecture based processors demonstrate up to a 200% temperature rise in neighboring cores that remain idle, underscoring the critical impact of spatial thermal interactions.The proposed solution establishes a "Proximity-Induced" thermal threshold for each compute core, dynamically governed by task priority. Prior to a core attaining its thermal limit, adjacent cores executing lower-priority tasks are proactively transitioned to reduced power states. This preemptive strategy mitigates premature thermal escalation in cores executing high-priority task, resulting in an observed increase of more than 30% in additional thermal headroom.
Fields of Study & Subjects
Parallel Computing and Optimization Techniques
Real-Time Systems Scheduling
Low-power high-performance VLSI design
Publication details
- Journal / source
- International Conference on Consumer Electronics
- Pages
- 1–6
- Publication date
- 3 February 2026
- Language
- EN