Domain Adaptation for CI/CD Workload Prediction: Transferring Seasonality Patterns Across Non-Overlapping Time Windows

Juan Camilo Escobar-Naranjo; Nestor Darío Duque-Mendez

doi:10.67294/7zhrs226

Authors

Juan Camilo Escobar-Naranjo National University of Colombia Author https://orcid.org/0009-0008-3262-1668
Nestor Darío Duque-Mendez National University of Colombia Author https://orcid.org/0000-0002-4608-281X

DOI:

https://doi.org/10.67294/7zhrs226

Keywords:

Transfer Learning, Domain Adaptation, Workload Prediction, Seasonality Patterns, Continuous Integration

Abstract

Machine learning models for workload prediction typically require training data from the same time period as deployment. In CI/CD environments, obtaining representative real-world training data is often difficult due to privacy concerns, data availability constraints, or the need to predict for newly established infrastructure. This paper presents a transfer learning methodology that enables transferring seasonality patterns and workload characteristics from real billing data collected during one-time period to real workflow execution logs from a completely non-overlapping period. The article approach combines two techniques: (1) Quantile Mapping to align duration distributions between the target- and source-domain data, preserving structural characteristics (shape, variability, skewness), and (2) Cross-Temporal Seasonality Transfer to map weekly activity patterns across time windows. Validated on GitHub Actions data spanning two distinct quarters (January-April 2023 billing data transferred to June-October 2023 real execution logs), our method achieved a strong correlation (r=0.8739, p=0.0101) for weekly patterns, explaining 76.4% of the variance in seasonality (R2=0.764), and aligned the duration distribution, reducing the CV error from 0.26 to 0.01, skewness from 1.06 to 0.03, and kurtosis from 2.33 to 0.07 relative to the source domain – all without overlapping temporal coverage between the two data sources.

References

Beloglazov, A., Buyya, R., Lee, Y. C., & Zomaya, A. (2011). A taxonomy and survey of energy-efficient data centers and cloud computing systems. In M. V. Zelkowitz (Ed.), Advances in Computers (Vol. 82, pp. 47–111). Elsevier. https://doi.org/10.1016/B978-0-12-385512-1.00003-7

Ben-David, S., Blitzer, J., Crammer, K., Kulesza, A., Pereira, F., & Vaughan, J. W. (2010). A theory of learning from different domains. Machine Learning, 79(1-2), 151–175. https://doi.org/10.1007/s10994-009-5152-4

Bisong, E., Tran, E., & Baysal, O. (2017). Built to last or built too fast? Evaluating prediction models for build times. 2017 IEEE/ACM 14th International Conference on Mining Software Repositories (MSR), 487–496. https://doi.org/10.1109/MSR.2017.14

Box, G. E. P., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: Forecasting and control (5th ed.). Wiley.

Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324

Calheiros, R. N., Masoumi, E., Ranjan, R., & Buyya, R. (2014). Workload prediction using ARIMA model and its impact on cloud applications' QoS. IEEE Transactions on Cloud Computing, 3(4), 449–458. https://doi.org/10.1109/TCC.2014.2350475

Cannon, A. J., Sobie, S. R., & Murdock, T. Q. (2015). Bias correction of GCM precipitation by quantile mapping: How well do methods preserve changes in quantiles and extremes? Journal of Climate, 28(17), 6938–6959. https://doi.org/10.1175/JCLI-D-14-00754.1

Chen, B., Chen, L., Zhang, C., & Peng, X. (2020). BuildFast: History-aware build outcome prediction for fast feedback and reduced cost in continuous integration. 2020 35th IEEE/ACM International Conference on Automated Software Engineering (ASE), 42–53. https://doi.org/10.1145/3324884.3416616

Claes, M., Mäntylä, M. V., Kuutila, M., & Adams, B. (2018). Do programmers work at night or during the weekend? Proceedings of the 40th International Conference on Software Engineering (ICSE ’18), 705–715. https://doi.org/10.1145/3180155.3180193

Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A seasonal-trend decomposition procedure based on loess. Journal of Official Statistics, 6(1), 3–73.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.

Dean, J., & Barroso, L. A. (2013). The tail at scale. Communications of the ACM, 56(2), 74–80. https://doi.org/10.1145/2408776.2408794

Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735

Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: Principles and practice (3rd ed.). OTexts. https://otexts.com/fpp3/

Jin, X., Park, Y., Maddix, D., Wang, H., & Wang, Y. (2022). Domain adaptation for time series forecasting via attention sharing. Proceedings of the 39th International Conference on Machine Learning (ICML), PMLR 162, 10280–10297.

Kinsman, T., Wessel, M., Mens, T., & Serebrenik, A. (2022). How do software developers use GitHub Actions to automate their workflows? Proceedings of the 2021 IEEE/ACM 18th International Conference on Mining Software Repositories (MSR), 420–431. https://doi.org/10.1109/MSR52588.2021.00054

Mao, H., Alizadeh, M., Menache, I., & Kandula, S. (2016). Resource management with deep reinforcement learning. Proceedings of the 15th ACM Workshop on Hot Topics in Networks (HotNets), 50–56. https://doi.org/10.1145/3005745.3005750

Maraun, D. (2013). Bias correction, quantile mapping, and downscaling: Revisiting the inflation issue. Journal of Climate, 26(6), 2137–2143. https://doi.org/10.1175/JCLI-D-12-00821.1

Moriconi, S., Janes, A., & Russo, B. (2024). GHALogs: A large-scale dataset of executed GitHub Actions workflows [Data set]. Zenodo. https://doi.org/10.5281/zenodo.10813696

Pan, S. J., & Yang, Q. (2010). A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359. https://doi.org/10.1109/TKDE.2009.191

Patki, N., Wedge, R., & Veeramachaneni, K. (2016). The synthetic data vault. Proceedings of the 2016 IEEE International Conference on Data Science and Advanced Analytics, 399–410. https://doi.org/10.1109/DSAA.2016.49

Pintye, P., Horváth, T., Balogh, A., & Charaf, H. (2024). Autoscaling GitHub Actions runners: Benefits and challenges. Proceedings of the 2024 IEEE International Conference on Cloud Computing, 145–152.

Rossi, F., Cardellini, V., & Presti, F. L. (2019). A review of autoscaling techniques for elastic applications in cloud environments. Journal of Grid Computing, 17(4), 699–726. https://doi.org/10.1007/s10723-019-09501-x

Saidani, I., Ouni, A., & Mkaouer, M. W. (2022). Improving the prediction of continuous integration build failures using deep learning. Automated Software Engineering, 29(1), 21. https://doi.org/10.1007/s10515-021-00319-5

Saxena, D., Kumar, J., Singh, A. K., & Schmid, S. (2023). Performance analysis of machine learning centered workload prediction models for cloud. arXiv preprint arXiv:2302.02452.

Subirats, J., & Guitart, J. (2015). Assessing and forecasting energy efficiency on cloud computing platforms. Future Generation Computer Systems, 45, 70–94. https://doi.org/10.1016/j.future.2014.11.008

United Nations. (2015). Transforming our world: The 2030 agenda for sustainable development (General Assembly resolution A/RES/70/1).

Vasilescu, B., Yu, Y., Wang, H., Devanbu, P., & Filkov, V. (2015). Quality and productivity outcomes relating to continuous integration in GitHub. Proceedings of the 2015 10th Joint Meeting on Foundations of Software Engineering (ESEC/FSE), 805–816. https://doi.org/10.1145/2786805.2786850

Weiss, K., Khoshgoftaar, T. M., & Wang, D. (2016). A survey of transfer learning. Journal of Big Data, 3(1), 1–40. https://doi.org/10.1186/s40537-016-0043-6

Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H., & He, Q. (2021). A comprehensive survey on transfer learning. Proceedings of the IEEE, 109(1), 43–76. https://doi.org/10.1109/JPROC.2020.3004555