ACM SOSP’26 Papers & Preprints | pchaigno
The SOSP’26 conference has published its list of accepted papers.<br>I was preparing to share that list when I noticed that many of the papers already have preprint versions available on arXiv or elsewhere.<br>This blog post therefore reproduces the list here, with all the preprint papers I could find.<br>Note some of the titles are probably not final; they sometimes use generic names (ex., “SystemX”) that sound like anonymized versions.<br>Similarly, the papers have probably changed since their preprint versions.
The bar plot above illustrates the variety of research areas covered.<br>The research areas are taken from last year’s schedule and I categorized the papers based on the titles only, so take with a grain of salt.<br>For papers touching multiple research areas, I took what I viewed as the main area1.
A Few GPUs, A Whole Lotta Scale: Faithful LLM Training Emulation with CrystalLLM
Shaoke Xi, ChonLam Lao (Alibaba Group), Boyi Jia (Shanghai Jiao Tong University and Alibaba Group), Jiaqi Gao, Zhipeng Zhang, Jiamin Cao (Alibaba Group), Brian Sutioso (Harvard University), Erci Xu (Shanghai Jiao Tong University), Minlan Yu (Harvard University), Kui Ren (Zhejiang University), Yong Li, Zhengping Qian, Ennan Zhai, Jingren Zhou (Alibaba Group)
Paper<br>Abstract
Large language model (LLM) training today runs on clusters spanning thousands of GPUs. While this scale enables rapid<br>model advances, developing, debugging, and performance-tuning, the training framework inevitably becomes complex and costly. This is because engineers often need to reproduce production behaviors to diagnose failures or evaluate optimizations, thereby demanding frequent and even exclusive access to production-scale clusters—which becomes increasingly hard given that the majority of GPUs are already committed to production workloads. Simulation relies on complex performance models that are difficult to maintain, and downscaled experiments often fail to capture scale-dependent behaviors.
We present PrismLLM to decouple large-scale execution from the need to access large clusters, enabling engineers to run and observe ranks of interest under faithful large-scale behavior using only a few GPUs. PrismLLM constructs a high-fidelity execution graph via a slicing-based approach that captures computation, communication, and dependencies of the target scale. Then, PrismLLM performs hybrid emulation where selected ranks execute the original program while the remaining are replayed as virtual participants.
Experiments on large-scale LLM training workloads show that PrismLLM accurately reproduces performance and memory behavior, achieving only 0.58% average error in iteration time and less than 0.01% error in peak GPU memory usage. PrismLLM can emulate clusters of up to 8192 GPUs using fewer than 1% of the physical GPUs required by the original deployment.
Ahead-of-time Analysis of Shell Program Effects
Lukas Lazarek, Evangelos Lamprou, George Kapetanakis, Eric Zhao, Zhiwen Zheng (Brown University), Michael Greenberg (Stevens Institute of Technology), Konstantinos Kallas (University of California, Los Angeles), Nikos Vasilakis (Brown University)
Paper<br>Abstract
The Unix shell remains a core system substrate across system administration, automation, and software development.<br>Shell programs, however, are prone to subtle, severe, and often irreversible effects that are difficult to predict. The challenge stems from the shell's unique execution model, its reliance on external computation and state, its highly dynamic expansion semantics, and the complex interactions<br>among these features. This paper presents SaSh, a system that statically analyzes shell programs to identify errors in<br>their execution before they occur. SaSh introduces an optimistic symbolic execution engine for shell programs that limits path explosion and focuses on high-impact failures. It tracks the effects of external commands over a filesystem model, and approximates shell word expansion using a tailored abstract domain. SaSh quickly identifies bugs even in large programs with a risk-directed exploration strategy, steering its analysis to program fragments likely to exhibit dangerous behavior. Applied to 61 buggy programs, including several high-profile disasters, SaSh identifies all but one instance of unwanted behavior with no false positives, going far beyond the current state-of-the-art. Furthermore, SaSh has already yielded 70 bug reports in 44 open-source projects such as PyTorch, the P4 Compiler, Kubernetes, and vLLM, including bugs that can lead to irreversible data loss.
Anchor: Mitigating Shallow Disruptions with Decoupled Memory
Haoyi Ma, Shiwei Gao (Tsinghua University), Youmin Chen (Shanghai Jiao Tong University), Junrong Huang, Youyou Lu, Jiwu Shu (Tsinghua University)
Axiom: Achieving Determinism in LLM Inference
Raja Gond (Microsoft Research), Aditya K Kamath (University of Washington), Ramachandran Ramjee, Ashish Panwar...