ut-osa / gpunet
GPUnet is a native GPU networking layer that provides a socket abstraction over Infiniband to GPU programs for NVIDIA GPUs.
☆92Updated 9 years ago
Related projects ⓘ
Alternatives and complementary repositories for gpunet
- GPUDirect Async support for IB Verbs☆90Updated 2 years ago
- ☆30Updated 7 years ago
- ☆40Updated 7 years ago
- Pointer-chasing memory benchmark (forked from Doug Pase's code).☆58Updated 10 years ago
- GPUfs - File system support for NVIDIA GPUs☆90Updated 5 years ago
- Donard: A PCIe Peer-2-Peer kernel patch and library that builds on top of NVM. Express. Also see https://github.com/sbates130272/linux-do…☆31Updated 8 years ago
- User-space Page Management☆104Updated 3 months ago
- ☆68Updated 8 years ago
- ☆63Updated 7 years ago
- OFI Programmer's Guide☆49Updated last year
- Linux Cross-Memory Attach☆88Updated 2 months ago
- HeteroSync is a benchmark suite for performing fine-grained synchronization on tightly coupled GPUs☆27Updated 2 months ago
- MapReduce for multi-core☆49Updated 11 years ago
- CUPTI GPU Profiler☆37Updated 5 years ago
- Parallel Memory Bandwidth Measurement / Benchmark Tool☆104Updated 2 years ago
- Enabling on-the-fly manipulations with LLVM IR code of CUDA sources☆101Updated last year
- Tensorflow is a computational library using data flow graphs for scalable machine learning, and Tensorflow-RDMA is the implementation ov…☆59Updated last year
- Infiniband verbs performance tests (fork of git://git.openfabrics.org/~grockah/perftest.git)☆16Updated 8 years ago
- FROZEN: the master branch has merged with the libfabric git repo☆31Updated 6 years ago
- Experiments evaluating preemption on the NVIDIA Pascal architecture☆18Updated 8 years ago
- A low-overhead tool to periodically collect system-wide hardware performance counters on Intel64 systems.☆31Updated 2 years ago
- NVIDIA GPU direct RDMA using SISCI API☆16Updated 6 years ago
- ☆32Updated 7 years ago
- Automatic virtualization of (general) accelerators.☆40Updated last year
- IX a protected dataplane operating system project☆127Updated 7 years ago
- NVM key-value store API lIbrary repository.☆88Updated 10 years ago
- Byte-Addressable File System☆19Updated 3 years ago
- Blaze runtime system that support efficient accelerator integration for big data.☆23Updated 7 years ago
- a multi-node fabric-attached memory manager that provides simple abstractions for accessing and allocating NVM from fabric-attached memor…☆9Updated 5 months ago
- Provides a set of benchmarks that can be used to measure the memory bandwidth performance of CPU's☆80Updated 7 months ago