# Characterizing Software Aging in GPU-Based LLM Serving Systems

arXiv: https://arxiv.org/abs/2606.11916 · HTML: https://mlsanity.com/paper/2606.11916/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 24 references that are in the map.
- First posted on arXiv 10 Jun 2026.

## Builds on (most cited first, 7 of 7)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention](https://mlsanity.com/paper/2405.04437/) (2024)
- [Jenga: Effective Memory Management for Serving LLM with Heterogeneity](https://mlsanity.com/paper/2503.18292/) (2025)
- [SYMPHONY: Improving Memory Management for LLM Inference Workloads](https://mlsanity.com/paper/2412.16434/) (2024)
- [Investigating Software Aging in LLM-Generated Software Systems](https://mlsanity.com/paper/2510.24188/) (2025)
- [Software Micro-Rejuvenation for Android Mobile Systems](https://mlsanity.com/paper/2112.06874/) (2021)
- [Adaptive Detection of Software Aging under Workload Shift](https://mlsanity.com/paper/2511.03103/) (2025)

## Abstract (arXiv)

This paper proposes an empirical methodology to study software aging in GPU-based LLM serving systems. Traditional aging studies focus on CPU-centric software with relatively regular workloads; LLM serving is different, spanning a Python host and a CUDA device, handling requests whose cost varies by orders of magnitude, and relying on rapidly evolving software stacks. We run a 216-hour campaign across six co-located deployments under identical stress conditions, monitor host, device, and client metrics in parallel, and apply a statistical pipeline that accounts for autocorrelation and multiple testing. Our results reveal statistically significant memory aging in all deployments, with leak rates strongly dependent on the serving runtime and deployment configuration. Beyond these findings, we provide a reproducible framework that opens a research direction at the intersection of the software aging and rejuvenation and LLM serving communities.