All publicationsAmir Noohi
Workshop2025

RMAI: Rethinking Memory for AI (Inference) In-Kernel Remote Shared Memory as a Software Alternative to CXL

Amir Noohi, Mostafa Derispour, Antonio Barbalace

Proceedings of the 5th Workshop on Machine Learning and Systems (EuroMLSys), pp. 122-131

PDFSlidesDOI
5 citations1122 downloads

Abstract

As AI models grow exponentially in size, memory has emerged as a critical bottleneck for inference at scale. While hardware solutions like Compute Express Link (CXL) promises to solve the problem of memory capacity and sharing, they require capital investment, and are not widely available. This paper presents RMAI, an in-kernel remote shared memory framework tailored for AI inference workloads, offering a transparent, scalable, and cost-effective software alternative to hardware-based memory expansion and sharing solutions.

BibTeX

@inproceedings{noohi2025rmai,
  author    = {Amir Noohi and Mostafa Derispour and Antonio Barbalace},
  title     = {{RMAI: Rethinking Memory for AI (Inference) In-Kernel Remote Shared Memory as a Software Alternative to CXL}},
  booktitle = {Proceedings of the 5th Workshop on Machine Learning and Systems (EuroMLSys)},
  year      = {2025},
  pages     = {122--131},
  doi       = {10.1145/3721146.3721954},
}