The NP-complete combinatorial optimization task of assigning offsets to a set of buffers with known sizes and lifetimes so as to minimize total memory usage is called dynamic storage allocation (DSA). Existing DSA implementations bypass the theoretical state-of-the-art algorithms in favor of either fast but wasteful heuristics, or memory-efficient approaches that do not scale beyond one thousand buffers. The "AI memory wall", combined with deep neural networks' static architecture, has reignited interest in DSA. We present idealloc, a low-fragmentation, high-performance DSA implementation designed for million-buffer instances. Evaluated on a novel suite of particularly hard benchmarks from several domains, idealloc ranks first against four production implementations in terms of a joint effectiveness/robustness criterion.
@article{arxiv.2504.04874,
title = {Futureproof Static Memory Planning},
author = {Christos Lamprakos and Panagiotis Xanthopoulos and Manolis Katsaragakis and Sotirios Xydis and Dimitrios Soudris and Francky Catthoor},
journal= {arXiv preprint arXiv:2504.04874},
year = {2025}
}