Reclamation Scheduling Under Mixed Enterprise Workloads: Trade-Offs Between Capacity Recovery and Tail Latency
Abstract
Reclamation tasks recover unused storage but can compete with production workloads for compute, metadata, and I/O resources. We evaluated five reclamation schedules using replayed traces from mixed database, analytics, virtual-desktop, and file-service workloads on a scale-out storage testbed. Policies ranged from immediate reclamation to load-aware deferral and fixed off-peak windows. The study measured recovered capacity, reclamation completion time, median latency, p99 latency, and backlog accumulation over 30-day trace replays. Immediate reclamation produced the fastest space recovery but increased p99 application latency by 17.8% during burst periods. Fixed overnight scheduling reduced interference but allowed reclamation backlogs to persist when overnight workload remained elevated. A load-aware policy using both queue depth and latency thresholds recovered 91% of eligible capacity within 24 hours while limiting p99 latency inflation to 4.1%. Performance differences were modest under steady workloads but substantial under concurrent analytics and backup activity. Reclamation should therefore be treated as a schedulable background service with explicit quality-of-service constraints. Adaptive scheduling offered the best balance between timely capacity recovery and protection of latency-sensitive applications.
References
1. Nazir M. Capacity reclamation and thin-provisioning efficiency in large-scale enterprise storage systems. International Journal of Business & Computational Sciences. 2021;1(1). Available from: https://ijbcs.org/index.php/IJBCS/article/view/2021P4
2. Patterson DA, Gibson G, Katz RH. A case for redundant arrays of inexpensive disks (RAID). In: Proceedings of the 1988 ACM SIGMOD International Conference on Management of Data. 1988. p. 109-116. doi:10.1145/50202.50214.
3. Meyer DT, Bolosky WJ. A study of practical deduplication. In: Proceedings of the 9th USENIX Conference on File and Storage Technologies. 2011. p. 1-13. Available from: https://www.usenix.org/legacy/events/fast11/tech/techAbstracts.html#Meyer
4. Linux kernel contributors. Thin provisioning [Internet]. [cited 2026 Sep 22]. Available from: https://docs.kernel.org/admin-guide/device-mapper/thin-provisioning.html
5. util-linux contributors. fstrim(8) - discard unused blocks on a mounted filesystem [Internet]. [cited 2026 Sep 22]. Available from: https://man7.org/linux/man-pages/man8/fstrim.8.html
6. Axboe J, fio contributors. fio documentation [Internet]. [cited 2026 Sep 22]. Available from: https://fio.readthedocs.io/en/latest/fio_doc.html
7. Mytkowicz T, Diwan A, Hauswirth M, Sweeney PF. Producing wrong data without doing anything obviously wrong. In: Proceedings of the 14th International Conference on Architectural Support for Programming Languages and Operating Systems. 2009. p. 265-276. doi:10.1145/1508244.1508275.
8. Kalibera T, Jones R. Rigorous benchmarking in reasonable time. In: Proceedings of the 2013 International Symposium on Memory Management. 2013. p. 63-74. doi:10.1145/2464157.2464160.
9. Dean J, Barroso LA. The tail at scale. Commun ACM. 2013;56(2):74-80. doi:10.1145/2408776.2408794.
10. Ghemawat S, Gobioff H, Leung ST. The Google file system. In: Proceedings of the 19th ACM Symposium on Operating Systems Principles. 2003. p. 29-43. doi:10.1145/945445.945450.
11. DeCandia G, Hastorun D, Jampani M, Kakulapati G, Lakshman A, Pilchin A, et al. Dynamo: Amazon's highly available key-value store. In: Proceedings of the 21st ACM Symposium on Operating Systems Principles. 2007. p. 205-220. doi:10.1145/1294261.1294281.
12. Swanson M, Bowen P, Phillips AW, Gallup D, Lynes D. Contingency planning guide for federal information systems. Gaithersburg (MD): National Institute of Standards and Technology. 2010;NIST SP 800-34 Rev. 1. doi:10.6028/NIST.SP.800-34r1.
13. Lakens D. Sample size justification. Collabra Psychol. 2022;8(1):33267. doi:10.1525/collabra.33267.
14. Nosek BA, Ebersole CR, DeHaven AC, Mellor DT. The preregistration revolution. Proc Natl Acad Sci U S A. 2018;115(11):2600-2606. doi:10.1073/pnas.1708274114.
15. Wilkinson MD, Dumontier M, Aalbersberg IJ, Appleton G, Axton M, Baak A, et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data. 2016;3:160018. doi:10.1038/sdata.2016.18.
16. Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Front Psychol. 2013;4:863. doi:10.3389/fpsyg.2013.00863.
17. Efron B. Bootstrap methods: another look at the jackknife. Ann Stat. 1979;7(1):1-26. doi:10.1214/aos/1176344552.
18. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol. 1995;57(1):289-300. doi:10.1111/j.2517-6161.1995.tb02031.x.
Published
Issue
Section
License
Authors retain copyright. Articles published under the Creative Commons Attribution 4.0 International licence (CC BY 4.0) may be shared and adapted for any purpose, including commercially, provided appropriate credit is given, a link to the licence is supplied, and changes are indicated. No additional legal or technological restrictions may be imposed. Third-party material is included only where its credit line permits. Licence: https://creativecommons.org/licenses/by/4.0/. Earlier publications remain subject to their stated licence and author agreements unless the rights holder authorizes a change.
