XaaS Containers: Performance-Portable Representation With Source and IR Containers

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Copik, Marcin, Alnuaimi, Eiman, Kamatar, Alok, Hayot-Sasson, Valerie, Madonna, Alberto, Gamblin, Todd, Chard, Kyle, Foster, Ian, Hoefler, Torsten
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911168670466048
author Copik, Marcin
Alnuaimi, Eiman
Kamatar, Alok
Hayot-Sasson, Valerie
Madonna, Alberto
Gamblin, Todd
Chard, Kyle
Foster, Ian
Hoefler, Torsten
author_facet Copik, Marcin
Alnuaimi, Eiman
Kamatar, Alok
Hayot-Sasson, Valerie
Madonna, Alberto
Gamblin, Todd
Chard, Kyle
Foster, Ian
Hoefler, Torsten
contents High-performance computing (HPC) systems and cloud data centers are converging, and containers are becoming the default method of portable software deployment. Yet, while containers simplify software management, they face significant performance challenges in HPC environments as they must sacrifice hardware-specific optimizations to achieve portability. Although HPC containers can use runtime hooks to access optimized MPI libraries and GPU devices, they are limited by application binary interface (ABI) compatibility and cannot overcome the effects of early-stage compilation decisions. Acceleration as a Service (XaaS) proposes a vision of performance-portable containers, where a containerized application should achieve peak performance across all HPC systems. We present a practical realization of this vision through Source and Intermediate Representation (IR) containers, where we delay performance-critical decisions until the target system specification is known. We analyze specialization mechanisms in HPC software and propose a new LLM-assisted method for automatic discovery of specializations. By examining the compilation pipeline, we develop a methodology to build containers optimized for target architectures at deployment time. Our prototype demonstrates that new XaaS containers combine the convenience of containerization with the performance benefits of system-specialized builds.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17914
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle XaaS Containers: Performance-Portable Representation With Source and IR Containers
Copik, Marcin
Alnuaimi, Eiman
Kamatar, Alok
Hayot-Sasson, Valerie
Madonna, Alberto
Gamblin, Todd
Chard, Kyle
Foster, Ian
Hoefler, Torsten
Distributed, Parallel, and Cluster Computing
High-performance computing (HPC) systems and cloud data centers are converging, and containers are becoming the default method of portable software deployment. Yet, while containers simplify software management, they face significant performance challenges in HPC environments as they must sacrifice hardware-specific optimizations to achieve portability. Although HPC containers can use runtime hooks to access optimized MPI libraries and GPU devices, they are limited by application binary interface (ABI) compatibility and cannot overcome the effects of early-stage compilation decisions. Acceleration as a Service (XaaS) proposes a vision of performance-portable containers, where a containerized application should achieve peak performance across all HPC systems. We present a practical realization of this vision through Source and Intermediate Representation (IR) containers, where we delay performance-critical decisions until the target system specification is known. We analyze specialization mechanisms in HPC software and propose a new LLM-assisted method for automatic discovery of specializations. By examining the compilation pipeline, we develop a methodology to build containers optimized for target architectures at deployment time. Our prototype demonstrates that new XaaS containers combine the convenience of containerization with the performance benefits of system-specialized builds.
title XaaS Containers: Performance-Portable Representation With Source and IR Containers
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2509.17914