Optimizing Microservice Performance: A Software Configuration Management Changes and Time Series Analysis Framework

Year 2026, Volume: 13 Issue: 1, 101 - 111, 31.01.2026

Fatih Bildirici , Savaş Takan , Keziban Seçkin Codal

Abstract

Modern microservice architectures demand agile configuration management to ensure optimal performance and reliability. However, the dynamic and distributed nature of these systems complicates the assessment of how configuration changes influence key performance indicators (KPIs) such as CPU utilization, memory consumption, response time, and error rates. In this study, we propose a time-series-driven approach for analyzing and optimizing the impact of configuration modifications in large-scale enterprise environments. Initially, we quantify the causal relationships between critical configuration parameters (cache size, thread pool size, and release complexity) and performance metrics using correlation and regression analyses. Subsequently, we employ time series modeling techniques (Prophet, ARIMA, and LSTM-based Autoencoder) to detect anomalies stemming from misconfigurations and traffic surges. To further enhance system performance, we integrate Bayesian Optimization and reinforcement learning methods for automated parameter tuning, demonstrating up to a 20–25% reduction in response times under realistic workloads. We also investigate deployment strategies—such as blue- green and canary releases and their interplay with rollback processes. Our findings underscore the significance of data-driven configuration management for microservices, offering actionable insights into achieving higher system stability, lower operational costs, and rapid recovery from performance anomalies.

Keywords

Software Configuration Management (SCM) , Time-Series Analysis , Performance Optimization , Anomaly Detection , Deep Learning

Yazılım Konfigürasyon Yönetimi Değişiklikleri ve Mikro Servis Performansını Optimize Etmek: Zaman Serisi Analizi Çerçevesi

Abstract

Modern mikro servis mimarileri, optimum performans ve güvenilirlik sağlamak için çevik konfigürasyon yönetimi gerektirir. Ancak bu sistemlerin dinamik ve dağıtık yapısı, konfigürasyon değişikliklerinin CPU kullanımı, bellek tüketimi, yanıt süresi ve hata oranları gibi temel performans göstergelerini (KPI) nasıl etkilediğinin değerlendirilmesini zorlaştırmaktadır. Bu çalışmada, büyük ölçekli kurumsal ortamlarda konfigürasyon değişikliklerinin etkisini analiz etmek ve optimize etmek için zaman serisi odaklı bir yaklaşım öneriyoruz. İlk olarak, korelasyon ve regresyon analizlerini kullanarak kritik yapılandırma parametreleri (önbellek boyutu, iş parçacığı havuzu boyutu ve sürüm karmaşıklığı) ile performans ölçümleri arasındaki nedensel ilişkileri ölçüyoruz. Daha sonra, yanlış konfigürasyonlardan ve trafik dalgalanmalarından kaynaklanan anormallikleri tespit etmek için zaman serisi modelleme tekniklerini (Prophet, ARIMA ve LSTM tabanlı Autoencoder) kullanıyoruz. Sistem performansını daha da artırmak için, otomatik parametre ayarı için Bayesian Optimizasyonu ve takviyeli öğrenme yöntemlerini entegre ediyoruz ve gerçekçi iş yükleri altında yanıt sürelerinde %20-25'e varan bir azalma gösteriyoruz. Ayrıca mavi-yeşil ve kanarya sürüm dağıtımı gibi dağıtım stratejilerini ve bunların geri alma süreçleriyle etkileşimini de araştırıyoruz. Bulgularımız, mikro servisler için veri odaklı konfigürasyon yönetiminin önemini vurgulamakta ve daha yüksek sistem kararlılığı, daha düşük operasyonel maliyetler ve performans anormalliklerinden hızlı ve efektif bir şekilde kurtarmak için içgsörüler ve öneriler sunmaktadır.

Keywords

Yazılım Konfigürasyon Yönetimi , Zaman Serisi Analizi , Performans Optimizasyonu , Anomali Tespiti , Derin Öğrenme

Ethical Statement

Herhangi bir etik çatışma bulunmamaktadır.

Supporting Institution

Bulunmamaktadır.

References

• [1] Newman S., Building Microservices, 2021, O’Reilly Media.
• [2] Schäffer E., Leibinger H., Stamm A., Brossog M., and Franke J., Configuration-based process and knowledge management by structuring the software landscape of global operating industrial enterprises with microservices, Procedia Manufacturing, 2018, 24, pp. 86–93.
• [3] Chen Y., Yan M., Yang D., Zhang X., and Wang Z., Deep attentive anomaly detection for microservice systems with multimodal time-series data, in 2022 IEEE International Conference on Web Services (ICWS), 2022, Barcelona, Spain, IEEE, pp. 373–378.
• [4] Pham L., Ha H., and Zhang H., Root cause analysis for microservice system based on causal inference: how far are we?, in Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering, 2024, Sacramento, USA, pp. 706–715.
• [5] Ning Y., Kazemi H., and Tahmasebi P., A comparative machine learning study for time series oil production forecasting: ARIMA, LSTM, and Prophet, Computers & Geosciences, 2022, 164, p. 105126.
• [6] Suryawan I.G.T., Putra I.K.N., Meliana P.M., and Sudipa I.G.I., Performance comparison of ARIMA, LSTM, and Prophet methods in sales forecasting, Sinkron: Jurnal dan Penelitian Teknik Informatika, 2024, 8(4), pp. 2410–2421.
• [7] Berlack H.R., Software Configuration Management, in Encyclopedia of Software Engineering, ed. J. Marciniak, 2002, John Wiley & Sons, Hoboken, NJ.
• [8] Kim G., Humble J., Debois P., Willis J., and Forsgren N., The DevOps Handbook: How to Create World-Class Agility, Reliability, & Security in Technology Organizations, 2021, IT Revolution.
• [9] Al-Debagy O. and Martinek P., A metrics framework for evaluating microservices architecture designs, Journal of Web Engineering, 2020, 19(3–4), pp. 341–370.
• [10] Zhu Y., Liu J., Guo M., Bao Y., Ma W., Liu Z., Song K., and Yang Y., BestConfig: tapping the performance potential of systems via automatic configuration tuning, in Proceedings of the 2017 Symposium on Cloud Computing, 2017, Santa Clara, USA, pp. 338–350.
• [11] Cryer J.D. and Chan K.S., Time Series Analysis: With Applications in R, 2nd ed., 2008, New York, NY, Springer.
• [12] Raja U., Hale D.P., and Hale J.E., Modeling software evolution defects: a time series approach, Journal of Software Maintenance and Evolution: Research and Practice, 2009, 21(1), pp. 49–71.
• [13] Rafferty G., Forecasting Time Series Data with Facebook Prophet: Build, Improve, and Optimize Time Series Forecasting Models Using the Advanced Forecasting Tool, 2021, Birmingham, UK, Packt Publishing.
• [14] Elsayed M.S., Le-Khac N.-A., Dev S., and Jurcut A.D., Network anomaly detection using LSTM based autoencoder, in Proceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks, 2020, Alicante, Spain, pp. 37–45.
• [15] Huang C.-Y., Performance analysis of software reliability growth models with testing-effort and change-point, Journal of Systems and Software, 2005, 76(2), pp. 181–194.
• [16] Farayola O.A., Hassan A.O., Adaramodu O.R., Fakeyede O.G., and Oladeinde M., Configuration management in the modern era: best practices, innovations, and challenges, Computer Science & IT Research Journal, 2023, 4(2), pp. 140–157.
• [17] Haug M., Olsen E.W., Cuevas G., and Rementeria S. (Eds.), Managing the Change: Software Configuration and Change Management: Software Best Practice 2, 2012, Springer Science & Business Media.
• [18] Zhang S. and Ernst M.D., Which configuration option should I change?, in Proceedings of the 36th International Conference on Software Engineering, 2014, Hyderabad, India, pp. 152–163.
• [19] Leon A., Software Configuration Management Handbook, 2015, Artech House.
• [20] Liu H.H., Software Performance and Scalability: A Quantitative Approach, 2011, John Wiley & Sons.
• [21] Gocheva-Ilieva S.G., Ivanov A.V., Voynikova D.S., and Boyadzhiev D.T., Time series analysis and forecasting for air pollution in small urban area: an SARIMA and factor analysis approach, Stochastic Environmental Research and Risk Assessment, 2014, 28, pp. 1045–1060.
• [22] Choraś M., Kozik R., Pawlicki M., Hołubowicz W., and Franch X., Software development metrics prediction using time series methods, in Computer Information Systems and Industrial Management, 18th International Conference (CISIM 2019), 2019, Belgrade, Serbia, Springer, pp. 311–323.
• [23] Ho A., Bui A.M.T., Nguyen P.T., and Di Salle A., Fusion of deep convolutional and LSTM recurrent neural networks for automated detection of code smells, in Proceedings of the 27th International Conference on Evaluation and Assessment in Software Engineering, 2023, Oulu, Finland, pp. 229– 234.
• [24] Le V.-H. and Zhang H., Log-based anomaly detection with deep learning: how far are we?, in Proceedings of the 44th International Conference on Software Engineering (ICSE ’22), 2022, Pittsburgh, USA, ACM, pp. 1356–1367.
• [25] Thill M., Konen W., Wang H., and Bäck T., Temporal convolutional autoencoder for unsupervised anomaly detection in time series, Applied Soft Computing, 2021, 112, p. 107751.
• [26] Pereira J.A., Acher M., Martin H., Jézéquel J.-M., Botterweck G., and Ventresque A., Learning software configuration spaces: A systematic literature review, Journal of Systems and Software, 2021, 182, p. 111044.
• [27] Kim J. and Choi S., Bayeso: A Bayesian optimization framework in Python, Journal of Open Source Software, 2023, 8(90), p. 5320.
• [28] Tang Y. and Chen X., Software development, configuration, monitoring, and management of artificial neural networks, Security and Communication Networks, 2022, 2022, 11 pages, DOI: 10.1155/2023/9864132.
• [29] Zhao G., Hassan S., Zou Y., Truong D., and Corbin T., Predicting performance anomalies in software systems at run-time, ACM Transactions on Software Engineering and Methodology, 2021, 30(3), pp. 1–33.
• [30] O’Connor R.V., Elger P., and Clarke P.M., Continuous software engineering—A microservices architecture perspective, Journal of Software: Evolution and Process, 2017, 29(11), p. e1866.