EUVIMEDEUVIMED entdecken
Crossref

Methodology for the Experimental Evaluation of Memory Allocation Strategies and Object Lifetime Models in C++ Systems

Maksim Martynov

Universal Library of Engineering Technology · 25.05.2026 · 3 (2) · 60-71

DOI: 10.70315/uloap.ulete.2026.0302011 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Universal Library of Engineering Technology, online erschienen am 25.05.2026. Verfasst von Maksim Martynov. Die Arbeit nennt 8 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The methodology considers the experimental evaluation of memory allocation strategies and object lifetime models in C++ systems as a reproducible research process grounded in a generative workload model, a phase-based execution structure, and an event-driven measurement scheme. The methodology’s relevance stems from the growing role of allocators in the performance of computing systems, where differences in memory placement and reclamation policies affect latency, fragmentation, peak resource consumption, and resilience under prolonged load. The aim of the work is to formalize a procedure for allocator comparison by elevating object lifetime to the status of a fundamental experimental parameter. The novelty of the approach lies in the integration of deterministic operation generation, explicit lifetime models, the RampUp, Steady, and BulkReclaim execution phases with churn modeled as a workload profile applied within them rather than as a separate structural phase, as well as a comparability protocol fixing the workload, seed, environment, the rules for accounting for non-applicable operations, and a capability-gated validation matrix that excludes invalid allocator × workload × lifetime combinations. The principal conclusions are that accounting for lifetime reveals the mechanisms of memory degradation arising from the intermingling of short-lived and long-lived objects. Configuring the methodology with arena separation by lifetime profile reduces the measured memory footprint overhead and stabilizes the live and peak reserved-byte counters tracked by the framework. Phase decomposition exposes the differences between warm-up, steady-state mode, churn load, and bulk memory return. The methodology will be useful for researchers, infrastructure software developers, performance engineers, and architects of C++ systems.

Zitieren

APA

Martynov, M. (2026). Methodology for the Experimental Evaluation of Memory Allocation Strategies and Object Lifetime Models in C++ Systems. Universal Library of Engineering Technology, 3(2), 60-71. https://doi.org/10.70315/uloap.ulete.2026.0302011

Vancouver

Martynov M. Methodology for the Experimental Evaluation of Memory Allocation Strategies and Object Lifetime Models in C++ Systems. Universal Library of Engineering Technology. 2026;3(2):60-71. doi:10.70315/uloap.ulete.2026.0302011

BibTeX
@article{Martynov2026,
  author = {Martynov, Maksim},
  title = {Methodology for the Experimental Evaluation of Memory Allocation Strategies and Object Lifetime Models in C++ Systems},
  journal = {Universal Library of Engineering Technology},
  year = {2026},
  volume = {3},
  number = {2},
  pages = {60--71},
  doi = {10.70315/uloap.ulete.2026.0302011}
}
Kostenlos starten →