General
Multilevel Graph Partitioning on GPU Supercomputers
arXiv cs.DC0 views
arXiv:2611.24264v1 Announce Type: new Abstract: Graph partitioning is a key enabler in computational science, with applications to sparse linear algebra, load-balanced parallel computation, and graph computations. In recent years, graphics processing units (GPUs) have become the preferred accelerator on the world's top supercomputers. Despite this, most existing distributed graph partitioners do not utilize GPUs. In this work, we develop a new multi-GPU distributed partitioner based on Jet, a prior single-GPU partitioner. We seek to optimize scalability without sacrificing quality. Our partitioner utilizes CUDA-aware MPI, one-sided communication routines, and an overall bulk-synchronous parallel structure. We demonstrate solution quality comparable to Jet. Our experiments show excellent weak scaling up to 256 GPUs, enabling partitioning of instances significantly larger than a single-GPU partitioner can handle.
Source:https://arxiv.org/abs/2611.24264PreviouseFAIR: Fair LLM Serving under Hourly Electricity CommitmentsNextScaleBench: An Extensible and Reproducible Experimental Platform for Kubernetes Autoscaling Research
Related posts
کارت گرافیک دست دوم در ۲۰۲۶: چک لیست کامل قبل از خرید
0dev.to
Watching mouse fight club teaches the critters how to fight
0Ars Technica
# I Wanted a Simple Video Editor. So I Built One with Gemini Hydrogen. I'm a neurocreative. I work with music, visuals, digital art and experiments that often end up becoming videos. For a long time
0dev.to
Making 8-page foldable zines from a single A4 sheet in the browser
0Hacker News
Show HN: Foolscap – a writing app that reads your book and never writes for you
0Hacker News
To make Moon landings easier, NASA is making life harder for Orion
0Ars Technica