Home // CLOUD COMPUTING 2026, The Seventeenth International Conference on Cloud Computing, GRIDs, and Virtualization // View article

A Multi-Resource Power Modeling Framework for Energy-Aware Cloud Simulation in CloudSim

Authors:
Awet Teklemariam Ghebrekidan
Raju Shrestha

Keywords: Cloud computing; energy efficiency; power modeling; multi-resource simulation; CloudSim; datacenter sustainability

Abstract:
The shift toward data-intensive and Artificial Intelligence (AI)-driven cloud workloads has rendered traditional Central Processing Unit (CPU)-centric power modeling increasingly insufficient for accurate datacenter energy estimation. While CloudSim is the de facto standard for evaluating energy-aware strategies, its native models predominantly focus on CPU utilization, effectively overlooking the non-trivial power contributions of auxiliary subsystems. This paper presents a modular multi-resource power modeling framework for CloudSim that integrates subsystem-level energy estimation into a unified simulation workflow. The framework introduces three critical extensions: (i) a high-resolution monitoring module capturing time-series utilization for CPU, Random Access Memory (RAM), bandwidth, and storage; (ii) modular, resource-specific power models that facilitate disaggregated energy attribution; and (iii) enhanced datacenter components supporting multi-dimensional resource capacities and tiered energy analytics.By providing a cohesive abstraction layer and unified Application Programming Interface (API), the framework enables consistent and extensible energy evaluation across heterogeneous infrastructure scales. Evaluation under high-intensity workloads indicates that traditional CPU-centric models obscure a non-trivial share of subsystem-level energy consumption, with non-CPU components representing approximately one-third of the total energy footprint in the evaluated stress-test scenario.

Pages: 42 to 47

Copyright: Copyright (c) IARIA, 2026

Publication date: April 19, 2026

Published in: conference

ISSN: 2308-4294

ISBN: 978-1-68558-373-6

Location: Lisbon, Portugal

Dates: from April 19, 2026 to April 23, 2026