rdma repositories
Trending repositories tagged rdma, ranked by trending score. Updated daily.
-
1
kvcache-ai/Mooncake
Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
GitHub · C++ · 7,088 stars · 1,507 forks · +22 stars in 24 hours · Trending score 2.80
-
2
uccl-project/uccl
UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)
GitHub · C++ · 1,588 stars · 199 forks · +2 stars in 24 hours · Trending score 1.12
-
3
Picom-code/pico-dgx-ds4
DeepSeek V4.1 Flash on 4x DGX Spark: 90/144 tok/s chat/code single-user, 376 tok/s at 16 distinct users, 1M context, images. Fork of blaesus/dsgx10.
GitHub · Cuda · 121 stars · 23 forks · 0 stars in 24 hours · Trending score 1.07
-
4
afdEP/afdEP-comm
Expert-parallel communication for Attention–FFN Disaggregation (AFD), built on DeepEP V2
GitHub · Python · 98 stars · 0 forks · 0 stars in 24 hours · Trending score 0.92
-
5
openucx/ucx
Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)
GitHub · C · 1,748 stars · 643 forks · +1 stars in 24 hours · Trending score 0.83
-
6
spidernet-io/spiderpool
Underlay and RDMA network solution of the Kubernetes, for bare metal, VM and any public cloud
GitHub · Go · 683 stars · 102 forks · +1 stars in 24 hours · Trending score 0.74
-
7
kindlingai/gr8
A from-scratch CUDA inference engine for 1-6 DGX Sparks (GB10): GLM-5.3, GLM-5.3-Flash and DeepSeek-V4.1-Flash
GitHub · Cuda · 83 stars · 16 forks · 0 stars in 24 hours · Trending score 0.70
-
8
ROCm/iris
AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming
GitHub · Python · 206 stars · 63 forks · 0 stars in 24 hours · Trending score 0.66
-
9
NVIDIA/daqiri
DAQIRI connects high bandwidth streaming sensor data to the NVIDIA software ecosystem and is part of NVIDIA's Holoscan platform for streaming, real time AI data processing
GitHub · C++ · 52 stars · 19 forks · 0 stars in 24 hours · Trending score 0.47
-
10
redis/rueidis
A fast Golang Redis client that supports Client Side Caching, Auto Pipelining, RDMA, etc.
GitHub · Go · 2,997 stars · 276 forks · 0 stars in 24 hours · Trending score 0.40
-
11
gpustack/gpustack-operator
Manage heterogeneous accelerators, vGPU slicing, shared KV cache, model delivery, and inference workloads.
GitHub · Go · 6 stars · 9 forks · 0 stars in 24 hours · Trending score 0.39
-
12
acryl-fabric/gpubase-fabric
JONATHAN GPUBASE Fabric — open-core Ethernet/RoCEv2 GPU cluster network fabric manager by ACRYL
GitHub · Makefile · 17 stars · 0 forks · 0 stars in 24 hours · Trending score 0.32
-
13
valkey-io/valkey-go
A fast Golang Valkey client that supports Client Side Caching, Auto Pipelining, and RMDA.
GitHub · Go · 696 stars · 96 forks · 0 stars in 24 hours · Trending score 0.29
-
14
unifabric-io/unifabric
Cloud native observability for the high-speed AI networks.
GitHub · Go · 103 stars · 6 forks · 0 stars in 24 hours · Trending score 0.23
-
15
rustfs/GPUCache
CPU/TCP shared KV-cache contracts for AI inference, with shard fencing and a mock RustFS cold tier. BlueField-3, RDMA, GDS, and real engine integration remain in development; no hardware performance claims.
GitHub · Rust · 17 stars · 0 forks · 0 stars in 24 hours · Trending score 0.14
-
16
OpenVisualCloud/Media-Transport-Library
A real-time media transport (DPDK, AF_XDP) stack for both raw and compressed video based on COTS hardware.
GitHub · C · 254 stars · 93 forks · 0 stars in 24 hours · Trending score 0.08
-
17
yuuki/rdma_exporter
Prometheus exporter for collection of RDMA (RoCE) NIC statistics from Linux hosts
GitHub · Go · 54 stars · 12 forks · 0 stars in 24 hours · Trending score 0.06
-
18
openlake-project/openlake
OpenLake is a high performance storage engine for efficient LLM inference and GPU Training
GitHub · Rust · 3,094 stars · 421 forks · 0 stars in 24 hours · Trending score 0.04
-
19
fpgasystems/Coyote
Framework providing operating system abstractions and a range of shared networking and memory services for common modern heterogeneous platforms.
GitHub · SystemVerilog · 468 stars · 135 forks · 0 stars in 24 hours · Trending score 0.00
-
20
uccl-project/rdmatop
htop-like TUI for real-time RDMA network monitoring.
GitHub · Rust · 123 stars · 10 forks · 0 stars in 24 hours · Trending score 0.00
-
21
Tele-AI/Fluxon
An AI-native distributed data plane built in Rust that supports high performance RPC, KV Cache, Message Queue, and File & Object Acceleration
GitHub · Rust · 126 stars · 6 forks · 0 stars in 24 hours · Trending score 0.00
-
22
b-ostrov/MelonDMA
Experimental PCIDriverKit driver for Mellanox ConnectX RDMA cards on macOS — RoCEv2 clustering between Mac Studios, no kernel extension required.
GitHub · C++ · 10 stars · 0 forks · 0 stars in 24 hours · Trending score 0.00
-
23
aws-samples/sample-pacer
Peer-Accelerated Cache with EFA Replication — a node-local, S3-compatible caching proxy for Amazon EKS, in Rust. Unmodified S3 clients (boto3, AWS CLI, s5cmd, PyTorch) read from their own node: RAM or NVMe locally, a peer's copy over EFA RDMA, a miss streamed from Amazon S3.
GitHub · Rust · 15 stars · 1 forks · 0 stars in 24 hours · Trending score 0.00
-
24
christopherowen/dgx-spark-networking
Switchless RoCE collectives, NCCL profiles and fabric tooling for DGX Spark inference recipes
GitHub · Python · 30 stars · 3 forks · 0 stars in 24 hours · Trending score 0.00