EnE-Rep: An Energy-Efficient Data Replication Strategy for Clouds

The rapid rise in the popularity of cloud computing can be attributed to its inherent advantages. However, its expanding infrastructure leads to higher energy consumption and increased network latency. Virtual machine consolidation (VMC) and dynamic power management (DPM) are popular methods to impr...

Full description

Saved in:
Bibliographic Details
Published in:Baltic Journal of Modern Computing 2024, Vol.12 (3), p.304-326
Main Author: Alghobiri, Mohammed
Format: Article
Language:English
Subjects:
Access time
Carbon
Cloud computing
Computer centers
Cost analysis
Data replication
Energy consumption
Energy efficiency
Industrial plant emissions
Internet
Network latency
Performance measurement
Power management
Resource management
Virtual environments
Workloads
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rapid rise in the popularity of cloud computing can be attributed to its inherent advantages. However, its expanding infrastructure leads to higher energy consumption and increased network latency. Virtual machine consolidation (VMC) and dynamic power management (DPM) are popular methods to improve energy efficiency. However, these energy-saving approaches are incompatible with data replication. Our approach in this study is called EnE-Rep, that categorizes cloud data center nodes based on workload and applies targeted strategies for each category. In addition, EnE-Rep leverages a robust collection of components including a load manager, energy monitor, and replicator for achieving energy-efficient data replication. Furthermore, intelligent placement decisions are made based on key factors like CPU utilization, server proximity, available bandwidth, and memory usage. Finally, CloudSim simulations validate the effectiveness of EnE-Rep, demonstrating significant reductions in energy consumption alongside improved performance metrics such as VM migration frequency, host shutdown rate, and data access time.
ISSN:2255-8942
2255-8950
DOI:10.22364/bjmc.2024.12.3.06