Full Text PDF
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Vol 10 No 5 (2026): Volume 10, Issue 5, May 2026 | Pages: 651-667
International Research Journal of Innovations in Engineering and Technology
OPEN ACCESS | Research Article | Published Date: 29-05-2026
Explainable AI for scheduling has advanced most in cloud, cluster, and production settings, where explanation is used to interpret learned policies, support operator trust, and improve policy tuning [1]–[4]. Yet these settings rarely face the timing constraints of OS-level CPU scheduling, where even small overheads can perturb dispatch behavior, wakeup latency, and deadline satisfaction [5]–[7]. This review examines the impact of XAI on CPU scheduler behavior and usability, with a focus on the unresolved problem of real-time XAI at OS timescales. The review synthesizes literature from five connected areas: explainability for learned schedulers, machine learning in OS scheduling, programmable Linux scheduling substrates, explainable schedulability analysis, and low-impact tracing for real-time diagnosis [5], [6], [8]–[10]. It argues that current work contains the ingredients for real-time XAI, but not yet an integrated solution that jointly evaluates explanation fidelity, timing safety, and debugging value under microsecond-scale overhead limits. A fixed-priority real-time scheduler is used as the anchor case study because it exposes the strongest contrast between learned priority selection and explainable timing evidence [9], [11], [12]. The analysis shows that the most viable form of XAI for CPU scheduling is not heavyweight local attribution in the kernel fast path, but a layered design combining lightweight runtime evidence, trace-based diagnosis, and certificate-style schedulability support [6], [9], [10]. The review concludes by proposing a metric framework for real-time XAI that ties latency, jitter, deadline misses, explanation fidelity, and debugging value together. This positions real-time XAI for CPU scheduling as a distinct research problem at the boundary of interpretability, operating systems, and real-time systems.
Explainable AI, CPU scheduling, Real-time systems, Operating systems, Fixed-priority scheduling, Schedulability analysis, Linux scheduling.
I. Janidu Chinthana, & Samantha Rajapaksha. (2026). Explainable AI for CPU Scheduling Under Real-Time Constraints. International Research Journal of Innovations in Engineering and Technology - IRJIET, 10(5), 651-667. Article DOI https://doi.org/10.47001/IRJIET/2026.105088
This work is licensed under Creative common Attribution Non Commercial 4.0 Internation Licence
B. Li, Z. Lan, and M. E. Papka, “Interpretable modeling of deep reinforcement learning driven scheduling,” in 2023 31st International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS), 2023, pp. 1–8.
D. Fischer, H. M. H¨usener, F. Grumbach, L. Vollenkemper, A. M¨uller, and P. Reusch, “Demystifying reinforcement learning in production scheduling via explainable ai,” 2024. [Online]. Available: https://arxiv.org/abs/2408.09841
P. H. Leung, “Explainable ai for cpu resource scheduling in cloud operating systems,” Frontiers in Interdisciplinary Applied Science, 2025.
J. Chen, S. S. Banerjee, Z. T. Kalbarczyk, and R. K. Iyer, “Machine learning for load balancing in the linux kernel,” in Proceedings of the 11th ACM SIGOPS Asia-Pacific Workshop on Systems, ser. APSys ’20. New York, NY, USA: Association for Computing Machinery, 2020, p. 67–74. [Online]. Available: https://doi.org/10.1145/3409963.3410492
D. B. De Oliveira, D. Casini, J. Lelli, and T. Cucinotta, “Timerlat: Real-time linux scheduling latency measurements, tracing, and analysis,” IEEE Transactions on Computers, vol. 74, no. 8, pp. 2608–2620, 2025.
S. Schuster, P. W¨agemann, P. Ulbrich, and W. Schr¨oder-Preikschat, “Proving real-time capability of generic operating systems by system aware timing analysis,” in 2019 IEEE Real-Time and Embedded Tech nology and Applications Symposium (RTAS), 2019, pp. 318–330.
J. T. Humphries, N. Natu, A. Chaugule, O. Weisse, B. Rhoden, J. Don, L. Rizzo, O. Rombakh, P. Turner, and C. Kozyrakis, “ghost: Fast & flexible user-space delegation of linux scheduling,” in Proceedings of the ACM SIGOPS 28th Symposium on Operating Systems Principles, ser. SOSP ’21. New York, NY, USA: Association for Computing Machinery, 2021, p. 588–604. [Online]. Available: https://doi.org/10.1145/3477132.3483542
M. Maida, S. Bozhko, and B. B. Brandenburg, “Foundational Response-Time Analysis as Explainable Evidence of Timeliness,” in 34th Euromicro Conference on Real-Time Systems (ECRTS 2022), ser. Leibniz International Proceedings in Informatics (LIPIcs), M. Maggio, Ed., vol. 231. Dagstuhl, Germany: Schloss Dagstuhl– Leibniz Zentrum f¨ur Informatik, 2022, pp. 19:1–19:25. [Online]. Available: https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2022.19
S. Baruah and P. Ekberg, “Towards Efficient Explainability of Schedulability Properties in Real-Time Systems,” in 35th Euromicro Conference on Real-Time Systems (ECRTS 2023), ser. Leibniz International Proceedings in Informatics (LIPIcs), A. V. Papadopoulos, Ed., vol. 262. Dagstuhl, Germany: Schloss Dagstuhl Leibniz-Zentrum f¨ur Informatik, 2023, pp. 2:1–2:20. [Online]. Available: https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2023.2
H. Lee, J. Lee, I. Yeom, and H. Woo, “Panda: Reinforcement learning based priority assignment for multi-processor real-time scheduling,” IEEE Access, vol. 8, pp. 185570–185583, 2020.
S. Lee, H. Baek, H. Woo, K. G. Shin, and J. Lee, “Ml for rt: Priority assignment using machine learning,” in 2021 IEEE 27th Real-Time and Embedded Technology and Applications Symposium (RTAS), 2021, pp. 118–130.
F. Alzetta, P. Giorgini, A. Najjar, M. I. Schumacher, D. Calvaresi, “In time explainability in multi-agent systems: Challenges, opportunities, and roadmap.” Explainable, Transparent Autonomous Agents and Multi Agent Systems, D. Calvaresi, A. Najjar, M. Winikoff, and K. Fr¨amling, Eds., Cham: Springer International Publishing, pp. 39–53., 2020.
G. Buttazzo, “Bridging ai with real-time systems: Technical perspective,” Commun. ACM, vol. 67, no. 2, p. 109, Jan. 2024. [Online]. Available: https://doi.org/10.1145/3631339
Buttazzo, G., “Toward predictable ai-based real-time systems,” Real Time Systems, vol. 61, no. 2, pp. 237–252, Jun. 2025.
L. P. Horstmann, J. L. C. Hoffmann, and A. A. Fr¨ohlich, “A framework to design and implement real-time multicore schedulers using machine learning,” in 2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), 2019, pp. 251–258.
H. Fingler, I. Tarte, H. Yu, A. Szekely, B. Hu, A. Akella, and C. J. Rossbach, “Towards a machine learning-assisted kernel with lake,” in Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2, ser. ASPLOS 2023. New York, NY, USA: Association for Computing Machinery, 2023, p. 846–861. [Online]. Available: https://doi.org/10.1145/3575693.3575697
S. Miller, A. Kumar, T. Vakharia, A. Chen, D. Zhuo, and T. Anderson, “Enoki: High velocity linux kernel scheduler development,” in Proceedings of the Nineteenth European Conference on Computer Systems, ser. EuroSys ’24. New York, NY, USA: Association for Computing Machinery, 2024, p. 962–980. [Online]. Available: https://doi.org/10.1145/3627703.3629569
S. P. Koneru, M. O. Bouraima, and R. Rahman, “Adaptive user-space scheduling in linux: A performance study of sched ext for real-time systems,” in 2025 10th International Conference on Communication and Electronics Systems (ICCES), 2025, pp. 920–924.
F. Rajotte and M. R. Dagenais, “Real-time linux analysis using low-impact tracer,” Advances in Computer Engineering, vol. 2014, no. 1, p. 173976, 2014. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1155/2014/173976
V.-A. Nicolas, M. Lallali, S. Rubini, and F. Singhoff, “Verification of scheduling properties based on execution traces,” 2018. [Online]. Available: https://api.semanticscholar.org/CorpusID:236783450
J. Yo and A. Imam Kistijantoro, “Analyzing fair share fairness of tasks in the linux completely fair scheduler using ebpf,” in 2023 10th International Conference on Advanced Informatics: Concept, Theory and Application (ICAICTA), 2023, pp. 1–6.
J. C. Saez, A. Pousa, R. Rodri´ ıguez-Rodri´ ıguez, F. Castro,M. Prieto Matias, “Pmctrack: Delivering performance monitoring counter support to the os scheduler,” The Computer Journal, vol. 60, no. 1, pp. 60–85, Jan. 2017.
M. Kulkarni and T. Kamble, “Integration of machine learning into operating systems: A survey,” 2020. [Online]. Available: https://api.semanticscholar.org/CorpusID:245261176
Y. Zhang and Y. Huang, “”learned”: Operating systems,” SIGOPS Oper. Syst. Rev., vol. 53, no. 1, p. 40–45, Jul. 2019. [Online]. Available: https://doi.org/10.1145/3352020.3352027
F. Erlenbusch and N. Stricker, “Explainable reinforcement learning in job-shop scheduling: A systematic literature review.” Procedia CIRP, vol. 132, pp. 25–30, 2025.
S. G. Gal´an, M. Seddiki, R. J. P. de Prado, E. M. Exp´osito, A. Marchewka, and N. R. Reyes, “Relevance of using interpretability indexes for the design of schedulers in cloud computing systems,” in 2020 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2020, pp. 1–8.
e. a. Alzetta, F., “In-time explainability in multi-agent systems: Chal lenges, opportunities, and roadmap.” Explainable, Transparent Au tonomous Agents and Multi-Agent Systems, 2020.
T. Qinan, G. Xing, L. Guilin, and L. Juncong, “Optimizing linux scheduling based on global runqueue with scx,” in 2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2024, pp. 1813–1819.
N. Yang, X. Shu, and C. Liang, “An improved linux priority scheduling method based on xgboost,” in 2024 20th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC FSKD), 2024, pp. 1–8.
X. Wang, S. Jia, Z. Huang, J. Cao, and M. Song, “Mixture-of schedulers: An adaptive scheduling agent as a learned router for expert policies,” 2025. [Online]. Available: https://arxiv.org/abs/2511.11628
S. Y. Kahu, “Kerneloracle: Predicting the linux scheduler’s next move with deep learning,” 2025. [Online]. Available: https://arxiv.org/abs/2505.15213
S. Miller, A. Kumar, T. Vakharia, T. Anderson, A. Chen, and D. Zhuo, “Agile development of linux schedulers with ekiben,” 2023. [Online]. Available: https://arxiv.org/abs/2306.15076
Y. Jia, K. Tian, Y. You, Y. Chen, and K. Chen, “Skyloft: A general high-efficient scheduling framework in user space,” in Proceedings of the ACM SIGOPS 30th Symposium on Operating Systems Principles, ser. SOSP ’24. New York, NY, USA: Association for Computing Machinery, 2024, p. 265–279. [Online]. Available: https://doi.org/10.1145/3694715.3695973
K. Kaffes, J. T. Humphries, D. Mazi`eres, and C. Kozyrakis, “Syrup: User-defined scheduling across the stack,” in Proceedings of the ACM SIGOPS 28th Symposium on Operating Systems Principles, ser. SOSP ’21. New York, NY, USA: Association for Computing Machinery, 2021, p. 605–620. [Online]. Available: https://doi.org/10.1145/3477132.3483548
T. T. Sung and B. Ryu, “Deep reinforcement learning for system-on chip: Myths and realities,” IEEE Access, vol. 10, pp. 98048–98064, 2022.
D. Saxena, J. Chen, S. Yadalam, Y. Ro, R. Dwivedula, E. H. Campbell, A. Akella, C. J. Rossbach, and M. Swift, “How i learned to stop worrying and love learned os policies,” in Proceedings of the 2025 Workshop on Hot Topics in Operating Systems, ser. HotOS ’25. New York, NY, USA: Association for Computing Machinery, 2025, p. 1–7. [Online]. Available: https://doi.org/10.1145/3713082.3730384
C. Nagesh, G. Sudha Gowd, Naidu Kiran Kumar, G. Pradeep Reddy, “Machine learning techniques to optimize cpu scheduling in real-time systems: A comprehensive review and analysis,” IJARSCT, pp. 381–388, Jun. 2024.
Z. Kong, Y. Yadlapalli, S. Bateni, J. Guo, and C. Liu, “Lintsˆ rt: A learning-driven testbed for intelligent scheduling in embedded systems,” 2020. [Online]. Available: https://arxiv.org/abs/2007.05136
R. Glaubius, T. Tidwell, C. Gill, and W. D. Smart, “Real-time scheduling via reinforcement learning,” 2012. [Online]. Available: https://arxiv.org/abs/1203.3481
M. Cˆot´e and M. R. Dagenais, “Problem detection in real-time systems by trace analysis,” Advances in Computer Engineering, vol. 2016, no. 1, p. 9467181, 2016. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1155/2016/9467181
A.C. Arpaci-Dusseau, R. H. Arpaci-Dusseau, M. Livny, and J. Meehean, “Towards transparent cpu scheduling,” 2011. [Online]. Available: https://api.semanticscholar.org/CorpusID:61493292