Design and Practice of Elastic Scaling Mechanism for Medical Cloud-Edge Collaborative Architecture

Authors

  • Zhengyang Qi University of California, Irvine, CA, 92697, US

DOI:

https://doi.org/10.63593/JIMR.2788-7022.2025.10.003

Keywords:

cloud-edge collaboration, elastic scaling, medical peak prediction, multi-tenant isolation, CLIA compliance, edge preprocessing, seasonal event model, cost optimization

Abstract

Medical test peaks can triple within a few minutes, while traditional threshold-based scaling lags by 10 minutes and incurs a 45% higher cost, resulting in report delays exceeding 6 hours. This study proposes an integrated mechanism of “edge preprocessing + cloud elasticity + prediction trigger.” The edge filters invalid data in real time and reports the load, while the cloud pool scales up within 5 minutes based on “CPU > 75%” or “Flu-Prophet seasonal prediction.” Docker NS ensures CLIA-compliant hard isolation for multi-tenants. The experiment, based on 41 million real orders, maintained 99.93% availability and stabilized report issuance time at 2.4 hours under a 3.2× peak on Black Friday, with a 22% reduction in cloud bills. This study is the first to embed medical seasonal events into a cloud-edge collaborative closed loop, achieving non-collapse during peaks, cost savings, and compliance for easy replication in grassroots medical clouds.

Downloads

Published

2025-11-17

How to Cite

Qi, Z. . (2025). Design and Practice of Elastic Scaling Mechanism for Medical Cloud-Edge Collaborative Architecture. ournal of nnovations in edical esearch, 4(5), 13–18. https://doi.org/10.63593/JIMR.2788-7022.2025.10.003

Issue

Section

Articles