# Futureproof Static Memory Planning

arXiv: https://arxiv.org/abs/2504.04874 · HTML: https://mlsanity.com/paper/2504.04874/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 29 references that are in the map.
- First posted on arXiv 7 Apr 2025.

## Builds on (most cited first, 5 of 5)

- [Efficient Memory Management for Deep Neural Net Inference](https://mlsanity.com/paper/2001.03288/) (2020)
- [Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers](https://mlsanity.com/paper/2408.04413/) (2024)
- [Profile-guided memory optimization for deep neural networks](https://mlsanity.com/paper/1804.10001/) (2018)
- [MEMO: Fine-grained Tensor Management For Ultra-long Context LLM Training](https://mlsanity.com/paper/2407.12117/) (2024)
- [The Unexpected Efficiency of Bin Packing Algorithms for Dynamic Storage Allocation in the Wild: An Intellectual Abstract](https://mlsanity.com/paper/2305.01497/) (2023)

## Abstract (arXiv)

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.