https://jacr.sciforce.org/JACR/issue/feed 2026-01-21T09:32:00+00:00 Dr. Suryakiran Navath, Ph. D., editor@sciforce.net Open Journal Systems <p>The<strong> International Journal of Cloud Computing and Supply Chain Management</strong> is a peer-reviewed journal dedicated to the exploration and advancement of robotics and artificial intelligence (AI). It serves as a platform for researchers, engineers, and practitioners to share innovative solutions, research findings, and cutting-edge technologies in the field.</p> <p>The journal emphasizes practical applications of robotics and AI, fostering collaboration between academia and industry to address global challenges.</p> https://jacr.sciforce.org/JACR/article/view/251 2026-01-21T09:32:00+00:00 Tirumala Rao Gundala Tirumalagundala7@gmail.com

<p>This research investigates performance inefficiencies in cloud-native microservice applications built with Java. Using machine learning algorithms, we analyze system metrics including request rate, response time, and CPU utilization to predict and optimize throughput in dynamic cloud environments.Introduction: Microservices architectures enable scalability and resilience but present performance engineering challenges. Java-based microservices face memory consumption, startup time, and garbage collection issues. Current literature lacks systematic performance-oriented analysis and optimization approaches for cloud-native microservice systems.Research Significance: Performance engineering for microservices remains underexplored despite its criticality. This research addresses data collection, monitoring, and diagnostic challenges across multiple system layers. Findings help optimize Java microservice deployments, reducing latency, improving resource utilization, and enhancing user experience in cloud environments.Methodology: We collected 100 performance observations from cloud-native Java microservices, measuring request rate (rps), average response time (ms), CPU utilization (%), and system throughput (rps). We applied Linear Regression and Random Forest Regression models to establish correlations between metrics and predict system throughput, evaluating model accuracy using standard error metrics.Results and Discussion: Descriptive statistics revealed significant performance variability: request rates ranged from 113 to 1999 rps (mean: 1072.58), response times from 24.31 to 789.77 ms (mean: 394.71), and CPU utilization from 10.43% to 93.78% (mean: 53.52%). Linear Regression achieved R² of 0.9577 on training data; Random Forest Regression significantly outperformed with R² of 0.9922, demonstrating superior predictive accuracy. Both models generalized reasonably to test data, with RFR exhibiting better robustness in handling extreme values.Future Scope: Investigate dynamic workload management and containerization optimization techniques for Java microservices.</p>

2025-12-27T00:00:00+00:00 Copyright (c) 2026 International Journal of Cloud Computing and Supply Chain Management