Paper Contents
Abstract
ABSTRACTAs data keeps growing at an incredible pace, machine learning (ML) has become an essential part of making sense of it all especially in big data environments. But running ML algorithms can be energy-intensive, which isn't ideal for sustainability or keeping systems efficient. In this work, introduce a practical framework designed to make ML more energy-efficient, specifically within Hadoop-based big data platforms. This approach uses smart techniques like trimming unnecessary parts of models (model pruning), running tasks in parallel, improving how data is stored and accessed, and scheduling resources more effectively. To test this, used well-known datasets like NSL-KDD and Amazon Reviews to run classification and clustering tasks across distributed systems. The results were promising energy consumption and processing time were both significantly reduced, all without sacrificing accuracy. This research moves us a step closer to greener computing and shows how ML can be both powerful and sustainable at scale.Keywords: Energy Efficient, ML, Hadoop, Big Data.
Copyright
Copyright © 2025 S.T. Pavithra Devi. This is an open access article distributed under the Creative Commons Attribution License.
