<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Rdma on Noureddine RAMDI</title><link>https://ramdi.fr/tags/rdma/</link><description>Recent content in Rdma on Noureddine RAMDI</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 23 May 2026 20:41:27 +0000</lastBuildDate><atom:link href="https://ramdi.fr/tags/rdma/index.xml" rel="self" type="application/rss+xml"/><item><title>DeepEP: Optimizing communication for large Mixture-of-Experts models with CUDA kernels</title><link>https://ramdi.fr/github-stars/deepep-optimizing-communication-for-large-mixture-of-experts-models-with-cuda-kernels/</link><pubDate>Sat, 02 May 2026 20:07:04 +0000</pubDate><guid>https://ramdi.fr/github-stars/deepep-optimizing-communication-for-large-mixture-of-experts-models-with-cuda-kernels/</guid><description>DeepEP is a CUDA-based communication library designed for Mixture-of-Experts models, delivering high-throughput GPU kernels with NVLink and RDMA support for efficient expert parallelism.</description></item></channel></rss>