Characterization and Forecasting the Workload of DRLAB Medical Database Server Based on the Shift-Wise and Machine Learning Approaches

Characterization & Forecasting the Workload of DRLAB Medical Database

Authors

  • Saleem Razzak Qureshi Institute of Mathematics and Computer Science, University of Sindh Jamshoro, Jamshoro 76080, Pakistan.
  • Aftab Ahmed Chandio Institute of Mathematics and Computer Science, University of Sindh Jamshoro, Jamshoro 76080, Pakistan. https://orcid.org/0000-0002-5752-0520
  • Qamar-ul-Nisa Chandio Government Degree Boys College Qasimabad, Education & Literacy Department, Government of Sindh, Hyderabad 71000, Pakistan.

Keywords:

Characterization, Database Workload, Machine Learning, Regression Analysis, Medical Lab, ICT, LUMHS, COVID

Abstract

The fundamental process to understand systems’ workload is to analyze its impact and outlining workload characterization for better understanding, it enables systems’ owners and policy makers to make decisions regarding policy management in order to improve system performance. Digital healthcare system is being modernized very fast at global level, especially medical laboratories require more advancements to match with standardization therefore efficient workload management of medical servers is needed to ensure reliable performance and scalability.  This research focuses on workload characterization of medical database web servers, specifically within Diagnostic and Research Laboratory (DRLAB) at Liaquat University of Medical and Health Sciences (LUMHS), Jamshoro, Sindh, Pakistan. The laboratory runs completely on ICT-based infrastructure. It offers lots of end-users to connect with, like patients, lab technicians, doctors, administrative staff, IT persons, and office workers, which raises concerns about systems’ scalability as end-users’ requests vary throughout the day. To assess the load of end-users on servers’ performance the approach is being used in this study is to analyze server access log data collected over seven-day period (4th to10th September 2020), comprising over 160,000 requests. We broke down the information into four six-hour shifts: midnight (12:01 AM to 6:00 AM), morning (6:01 AM to 12:00 PM), noon (12:01 PM to 6:00 PM), and evening (6:01 PM to 12:00 AM). In this way, the status of different time intervals in aspect to rush time may be observed. Furthermore, based of four observations, the machine learning Regression analysis techniques applied to comparatively analysis. Moreover, the results will be helpful to scheme up the database performance policy. The policy makers/stakeholders mitigate the issue by figuring out the analyzed data/statistics for the future planning.

Author Biography

Aftab Ahmed Chandio, Institute of Mathematics and Computer Science, University of Sindh Jamshoro, Jamshoro 76080, Pakistan.

I am currently working as Associate Professor of Institute of Mathematics and Computer Science in University of Sindh, Jamshoro, Pakistan. I received my PhD, doctor of engineering degree in Computer Applied Technology from University of Chinese Academy of Sciences, Beijing, China and engaged in PhD research in Centre for Cloud Computing Research of Shenzhen Institutes of Advanced Technology (SIAT) of Chinese Academy of Sciences (CAS), and Shenzhen Cloud Computer Centre of National Supercomputing Centre Shenzhen, China, from July 2011 to January 2016, under the supervision of Dr. Cheng-Zhong Xu, IEEE Fellow, professor of Wayne State University Detroit USA. I obtained my BS (Hons) degree in Computer Science from University of Sindh, Jamshoro Pakistan, from January 2003 to December 2006. My research interests include resource management, job scheduling strategies, energy efficiency, and workload characterization for performance optimization of distributed systems such as cloud computing, and location-based services i.e., map-matching strategy for GPS trajectories. My research work appears in over 18 publications in journals and conferences, including impact factor journals CLUSCOMP (JCR-Q1), SUSCOM (JCR-Q2) and FITEE (JCR-Q2), ZTE COMM, IEEE/ACM ISPA/TRUSTCOM, IEEE ICARCV, IEEE SCOReD, IEEE WORLD S4, IEEE WCICA, IOV in Springer?s LNCS, INTAP in Springer?s CCIS and IMECS in IAENG?s LNECS. I served as an invited reviewer in several journals and conferences, including IEEE Trans. on Cloud Computing, CLUSCOMP, MONET, JPDC, IPCCC, IEEE CloudCom, IEEE ITSC and IEEE ICVES. I also served as a Session Chair in IEEE SCOReD 2017 in Putrajaya Malaysia, IEEE ISPA/TRUSTCOM 2013 in Melbourne Australia and Springer's 2nd IOV 2015 in Chengdu China. I was a recipient of the ?Best Researcher Award 2019? of University of Sindh Jamshoro Pakistan, the ?Best Paper Award? of 15th IEEE SCOReD 2017 Malaysia, the ?Excellence Performance Award? as the Volunteer of IEEE/ACM CCGrid 2015 Shenzhen China and the ?Dean Merit Scholarship? awards of SIAT CAS for 2012 as well as for 2015. I have ?STOOD THIRD? in BS (Hons) in Computer Science.

Downloads

Published

2026-04-30

Issue

Vol. 7 No. 1 (2026): The Sciencetech, Jan-Mar 2026

Section

Articles

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.