Formalization of Experimental Methods for Evaluating Memory Allocators, Taking into Account the Semantics of the Object Lifecycle
Keywords:
memory allocator, benchmarking methodology, object lifetime, lifecycle semantics, workload characterizationAbstract
Memory allocator evaluation still lacks a stable methodological basis, even though allocator behavior depends on workload heterogeneity, concurrency patterns, and the temporal semantics of allocated objects. This article addresses that gap through a formal analytical framework that treats object lifetime as an explicit experimental variable instead of a residual statistic. The study systematizes recent allocator research, identifies points at which benchmark practice loses methodological precision, and proposes a phase-based evaluation model that links allocation events with lifecycle classes, workload transitions, and metric interpretation. The presented materials include 11 recently peer-reviewed sources on allocator characterization, lifetime profiling, semantics-aware allocation, tiering, and benchmark methodology. Comparative analysis, conceptual synthesis, typologization, and analytical generalization shape the methodological basis. The analytical section derives a lifecycle-centered evaluation schema, distinguishes workload classes that require different observables, and formulates reporting rules for reproducible allocator assessment in server workloads, managed runtimes, heterogeneous-memory systems, and application-specific environments.
References
[1]. Zhou, Z., Gogte, V., Vaish, N., Kennelly, C., Xia, P., Kanev, S., Moseley, T., Delimitrou, C., & Ranganathan, P. (2024). Characterizing a memory allocator at warehouse scale. Association for Computing Machinery. https://doi.org/10.1145/3620666.3651350
[2]. Upadhyay, S., & Venkat, A. (2026). Shiny objects: Object-centric characterization of Chromium. ACM Transactions, 10(1). https://doi.org/10.1145/3788102
[3]. Maas, M., Andersen, D. G., Isard, M., Javanmard, M. M., McKinley, K. S., & Raffel, C. (2024). Combining machine learning and lifetime-based resource management for memory allocation and beyond. Communications of the ACM, 67(4). https://doi.org/10.1145/3611018
[4]. Jordan Montaño, S., Polito, G., Ducasse, S., & Tesone, P. (2024). Evaluating finalization-based object lifetime profiling. Association for Computing Machinery. https://doi.org/10.1145/3652024.3665514
[5]. Wang, R., Xu, M., & Asokan, N. (2024). SeMalloc: Semantics-informed memory allocator. Association for Computing Machinery. https://doi.org/10.1145/3658644.3670363
[6]. Blackburn, S. M., Cai, Z., Chen, R., Yang, X., Zhang, J., & Zigman, J. (2025). Rethinking Java performance analysis. Association for Computing Machinery. https://doi.org/10.1145/3669940.3707217
[7]. Sareen, K., Blackburn, S. M., Hamouda, S. S., & Gidra, L. (2024). Memory management on mobile devices. Association for Computing Machinery. https://doi.org/10.1145/3652024.3665510
[8]. Li, R., & Yadwadkar, N. (2025). Old is gold: Optimizing single-threaded applications with Exgen-Malloc. arXiv. https://doi.org/10.48550/arXiv.2510.10219
[9]. Filardo, N. W., & Parkinson, M. J. (2024). BatchIt: Optimizing message-passing allocators for producer-consumer workloads: An intellectual abstract. Association for Computing Machinery. https://doi.org/10.1145/3652024.3665506
[10]. Kammerdiener, B., McMichael, J. Z., Jantz, M., Doshi, K., & Jones, T. (2025). Flexible and effective object tiering for heterogeneous memory systems. ACM Transactions, 22(1). https://doi.org/10.1145/3708540
[11]. Dang, Z., He, S., Zhang, X., Hong, P., Li, Z., Chen, X., Song, H., Sun, X.-H., & Chen, G. (2024). PMAlloc: A holistic approach to improving persistent memory allocation. ACM Transactions, 42(3–4). https://doi.org/10.1145/3643886
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Maksim Martynov

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who submit papers with this journal agree to the following terms.