Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Performance Measurements in the AI-Centric Computing Continuum Systems

Praveen Kumar Donta, Qiyang Zhang, Schahram Dustdar

TL;DR

The paper addresses the need for performance evaluation in AI-centric distributed computing continuums (DCC) comprising cloud, edge, and IoT layers. It surveys traditional metrics across computing, network, and user levels and argues they fall short for GenAI/LLM workloads. To address this gap, it proposes novel metrics—$CO_2$ emissions, heat dissipation, bottleneck fairness, observability, adaptivity quotient, and equilibrium maintenance—along with selection criteria to better capture sustainability, energy efficiency, adaptability, and transparency. The work aims to guide researchers and practitioners toward holistic, forward-looking measurement practices that support efficient, robust, and sustainable AI-driven DCC deployments.

Abstract

Over the Eight decades, computing paradigms have shifted from large, centralized systems to compact, distributed architectures, leading to the rise of the Distributed Computing Continuum (DCC). In this model, multiple layers such as cloud, edge, Internet of Things (IoT), and mobile platforms work together to support a wide range of applications. Recently, the emergence of Generative AI and large language models has further intensified the demand for computational resources across this continuum. Although traditional performance metrics have provided a solid foundation, they need to be revisited and expanded to keep pace with changing computational demands and application requirements. Accurate performance measurements benefit both system designers and users by supporting improvements in efficiency and promoting alignment with system goals. In this context, we review commonly used metrics in DCC and IoT environments. We also discuss emerging performance dimensions that address evolving computing needs, such as sustainability, energy efficiency, and system observability. We also outline criteria and considerations for selecting appropriate metrics, aiming to inspire future research and development in this critical area.

Performance Measurements in the AI-Centric Computing Continuum Systems

TL;DR

The paper addresses the need for performance evaluation in AI-centric distributed computing continuums (DCC) comprising cloud, edge, and IoT layers. It surveys traditional metrics across computing, network, and user levels and argues they fall short for GenAI/LLM workloads. To address this gap, it proposes novel metrics— emissions, heat dissipation, bottleneck fairness, observability, adaptivity quotient, and equilibrium maintenance—along with selection criteria to better capture sustainability, energy efficiency, adaptability, and transparency. The work aims to guide researchers and practitioners toward holistic, forward-looking measurement practices that support efficient, robust, and sustainable AI-driven DCC deployments.

Abstract

Over the Eight decades, computing paradigms have shifted from large, centralized systems to compact, distributed architectures, leading to the rise of the Distributed Computing Continuum (DCC). In this model, multiple layers such as cloud, edge, Internet of Things (IoT), and mobile platforms work together to support a wide range of applications. Recently, the emergence of Generative AI and large language models has further intensified the demand for computational resources across this continuum. Although traditional performance metrics have provided a solid foundation, they need to be revisited and expanded to keep pace with changing computational demands and application requirements. Accurate performance measurements benefit both system designers and users by supporting improvements in efficiency and promoting alignment with system goals. In this context, we review commonly used metrics in DCC and IoT environments. We also discuss emerging performance dimensions that address evolving computing needs, such as sustainability, energy efficiency, and system observability. We also outline criteria and considerations for selecting appropriate metrics, aiming to inspire future research and development in this critical area.

Paper Structure

This paper contains 18 sections, 6 equations, 2 figures.

Table of Contents

  1. Introduction
  2. Importance of Performance Measurements
  3. Traditional Performance Measurements
  4. Common goals of performance metrics in DCC
  5. Commonly used Performance Measurements
  6. Computing-level Performance Measurements
  7. Network-level Performance Measurement
  8. Application or User-level Performance Measurement
  9. Novel Performance Measurements
  10. Metrics
  11. $CO_2$ emissions
  12. Heat Dissipation
  13. Bottlenecks
  14. Measure Observability
  15. Adaptivity Quotient
  16. ...and 3 more sections

Figures (2)

  • Figure 1: Conceptual Framework of DCC Architecture
  • Figure 2: A taxonomy of commonly used performance measurements in DCC and IoT Networks