rdma
Here are 59 public repositories matching this topic...
GPU enabled OpenSHMEM using Vulkan
-
Updated
Jun 1, 2026 - C++
A zero-kernel, RDMA-first data plane that eliminates the AI "memory wall" and the I/O bottlenecks of the Linux Virtual File System (VFS). It bridges physical GPU VRAM, local NVMe, and CXL 3.1 Global Integrated Memory into a single, lock-free Silicon Data Substrate.
-
Updated
Apr 7, 2026 - C++
⚡️ A high-performance disaggregated memory pool built with C++20 coroutines and 1-sided RDMA.
-
Updated
Apr 2, 2026 - C++
Kdsync is a tool that allows ordered synchronization between Apache Kafka clusters. It internally uses Derecho library in order to achieve it.
-
Updated
Sep 22, 2021 - C++
Implement GPU collective communication using a custom PyTorch c10d backend built on libibverbs and softRoCE for distributed model training.
-
Updated
Aug 23, 2026 - C++
[bakup] An RDMA- & PMem-aware distributed KV storage system. (The original repo is on Gitee under the same account and project name.)
-
Updated
Jun 9, 2022 - C++
-
Updated
Sep 8, 2020 - C++
From-scratch RDMA-based PyTorch backend in C++; trained a char-level GPT on TinyShakespeare via DDP through code I wrote.
-
Updated
May 1, 2026 - C++
An extensible mechanism for pooling memory among remote machines.
-
Updated
Jul 1, 2021 - C++
This is a (modern) implementation of our paper Designing Distributed Tree-based Index Structures for Fast RDMA-capable Networks
-
Updated
Oct 8, 2023 - C++
Two scalable join implementations using MPI.
-
Updated
Jul 13, 2024 - C++
-
Updated
Jan 16, 2026 - C++
A micro-benchmark to measure RDMA READ and RDMA WRITE throughput (IOPS).
-
Updated
Jun 24, 2022 - C++
Improve this page
Add a description, image, and links to the rdma topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the rdma topic, visit your repo's landing page and select "manage topics."