Sign inSign up

fogesque/inflare-sdk

By fogesque

Updated 3 months ago

Development image with the inflare library

Image
Networking
0

2.1K

fogesque/inflare-sdk repository overview

fogesque/inflare-sdk

Development image with the Inflare RDMA streaming library installed.

Inflare is a C++ library for high-performance RDMA data streaming over IB Verbs, with optional CUDA / GPUDirect RDMA support for GPU memory streaming.

Tags

TagBasePurpose
basefogesque/rdma-cpp-dev-base:cpuCPU / IB Verbs Inflare SDK
gpunetiofogesque/rdma-cpp-dev-base:gpuCUDA-enabled Inflare SDK

Included

  • Inflare public headers
  • Static Inflare library
  • CMake package config
  • vcpkg-installed dependencies required by find_dependency()
  • RDMA and system inspection tools inherited from rdma-cpp-dev-base

CMake Usage

find_package(inflare CONFIG REQUIRED)
target_link_libraries(app PRIVATE inflare::inflare)

The image sets CMAKE_PREFIX_PATH so CMake can find Inflare and its packaged dependencies.

Example

docker run --rm -it \
    --network=host \
    --device=/dev/infiniband \
    --cap-add=IPC_LOCK \
    --ulimit memlock=-1 \
    -v "$PWD:/workspace/project" \
    fogesque/inflare-sdk:base

GPU builds should use the gpunetio tag and add:

--gpus all

Diagnostics

Run inside the container:

inflare-rdma-env

This checks visible RDMA devices, verbs providers, link state, PCI/NUMA information, and CUDA/NVIDIA state when available.

Host Requirements

The host must expose RDMA devices with --device=/dev/infiniband and allow memory locking for RDMA registration. GPU mode requires NVIDIA Container Toolkit, a compatible NVIDIA driver, CUDA-capable GPU, and GPUDirect RDMA host support where required.

Tag summary

Content type

Image

Digest

sha256:5a3f7dc27

Size

3.9 GB

Last updated

3 months ago

docker pull fogesque/inflare-sdk:0.3.3-gpunetio