Authors: Ismet Dagli, Amid Morshedlou, Jamal Rostami, Mehmet E. Belviranli

Published on: February 07, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.04522

Summary

• What is new: Introduces H-EYE, a new method for managing resources across the entire edge-cloud continuum with a multi-layer, graph-based hardware representation and a modular performance modeling interface.
• Why this is important: The need for better utilization of computational resources across different tiers of edge-cloud systems, which are currently managed in isolation.
• What the research proposes: A universal approach named H-EYE that captures diverse computational characteristics and manages task assignments across all tiers, using a novel orchestrator mechanism for mapping tasks to resources.
• Results: Demonstrates improvements in latency by up to 47% over baseline methods with less than 2% scheduling overhead in deployed edge-server systems.

Technical Details

Technological frameworks used: H-EYE

Models used: Multi-layer graph-based hardware representation, modular performance modeling interface, novel orchestrator mechanism

Data used: Not specified

Potential Impact

Companies providing cloud, edge computing services, and distributed computing solutions could benefit or be disrupted.

Want to implement this idea in a business?

We have generated a startup concept here: OptiEdge.