The Drift-Aware Uncertainty-Calibrated SLA Verification for Cloud Infrastructure Reliability
Keywords:
active probing, cloud computing, concept drift, conformal prediction, infrastructure as a service, reliabilityAbstract
Cloud customers commonly assess infrastructure reliability through short benchmark runs or provider-level availability reports. Such measurements are useful but can become misleading when workload mix, hardware placement, contention, software configuration, or network conditions change after the initial test. This paper proposes DriftGuard, a drift-aware and uncertainty-calibrated framework for independent verification of infrastructure-as-a-service performance commitments. The framework combines low-impact active probes with passive telemetry, aligns observations to explicit service-level objectives, detects distribution change, and produces reliability intervals rather than a single static score. A change detector monitors multivariate residuals between promised and observed service indicators. When drift or uncertainty rises, an adaptive probing policy selects the next test by expected information gain, operational risk, and monetary cost. When conditions remain stable, the policy reduces probe frequency to limit overhead. A calibrated prediction layer estimates whether throughput, latency, compute, memory, storage, and availability objectives will be satisfied over a defined verification horizon. Explanations identify the indicators and change events that drive each decision. The article specifies the mathematical model, adaptive algorithm, architecture, security controls, reproducible test protocol, baselines, ablations, and reporting requirements. It intentionally makes no unverified performance claim; numerical findings must be produced by executing the supplied protocol. The framework advances beyond fixed weighted scoring by treating cloud reliability as a time-varying, uncertain, and auditable property.
Downloads
References
[1] X. Wen and Y. Zheng, “The application of artificial intelligence technology in cloud computing environment resources,” J. Web Eng., vol. 20, no. 6, pp. 1853–1866, 2021, doi: 10.13052/jwe1540-9589.2067.
[2] P. Mell and T. Grance, The NIST Definition of Cloud Computing, NIST Special Publication 800-145, Sep. 2011.
[3] M. Armbrust et al., “A view of cloud computing,” Commun. ACM, vol. 53, no. 4, pp. 50–58, Apr. 2010.
[4] ISO/IEC 19086-1:2016, Cloud Computing Service Level Agreement Framework Part 1 Overview and Concepts, 2016.
[5] B. Beyer, C. Jones, J. Petoff, and N. R. Murphy, Eds., Site Reliability Engineering. Sebastopol, CA, USA: O’Reilly Media, 2016.
[6] J. Gama, I. Žliobaitė, A. Bifet, M. Pechenizkiy, and A. Bouchachia, “A survey on concept drift adaptation,” ACM Comput. Surv., vol. 46, no. 4, Art. no. 44, 2014.
[7] A. Bifet and R. Gavaldà, “Learning from time-changing data with adaptive windowing,” in Proc. SIAM Int. Conf. Data Mining, 2007, pp. 443–448.
[8] V. Vovk, A. Gammerman, and G. Shafer, Algorithmic Learning in a Random World. New York, NY, USA: Springer, 2005.
[9] A. N. Angelopoulos and S. Bates, “A gentle introduction to conformal prediction and distribution-free uncertainty quantification,” Found. Trends Mach. Learn., vol. 16, no. 4, pp. 494–591, 2023.
[10] M. T. Ribeiro, S. Singh, and C. Guestrin, “Why should I trust you? Explaining the predictions of any classifier,” in Proc. ACM SIGKDD, 2016, pp. 1135–1144.
[11] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Adv. Neural Inf. Process. Syst., vol. 30, 2017.
[12] V. Chandola, A. Banerjee, and V. Kumar, “Anomaly detection: A survey,” ACM Comput. Surv., vol. 41, no. 3, Art. no. 15, 2009.
[13] E. S. Page, “Continuous inspection schemes,” Biometrika, vol. 41, nos. 1–2, pp. 100–115, 1954.
[14] L. Breiman, “Random forests,” Mach. Learn., vol. 45, pp. 5–32, 2001.
[15] D. Sculley et al., “Hidden technical debt in machine learning systems,” in Adv. Neural Inf. Process. Syst., vol. 28, 2015.
[16] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, and I. Brandic, “Cloud computing and emerging IT platforms,” Future Gener. Comput. Syst., vol. 25, no. 6, pp. 599–616, 2009.
[17] L. A. Barroso, U. Hölzle, and P. Ranganathan, The Datacenter as a Computer, 3rd ed. Cham, Switzerland: Springer, 2018.
[18] B. R. Kandukuri, R. Paturi V., and A. Rakshit, “Cloud security issues,” in Proc. IEEE Int. Conf. Services Computing, 2009, pp. 517–520.
[19] R. Kohavi et al., Trustworthy Online Controlled Experiments. Cambridge, U.K.: Cambridge Univ. Press, 2020.
[20] A. Dean and D. Voss, Design and Analysis of Experiments, 2nd ed. New York, NY, USA: Springer, 2017.
[21] D. J. C. MacKay, Information Theory Inference and Learning Algorithms. Cambridge, U.K.: Cambridge Univ. Press, 2003.
[22] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed. New York, NY, USA: Springer, 2009.
[23] NIST, Guide for Conducting Risk Assessments, NIST Special Publication 800-30 Rev. 1, Sep. 2012.
[24] NIST, Security and Privacy Controls for Information Systems and Organizations, NIST Special Publication 800-53 Rev. 5, Sep. 2020.
