eth-cscs / Tiled-MMLinks
Matrix multiplication on GPUs for matrices stored on a CPU. Similar to cublasXt, but ported to both NVIDIA and AMD GPUs.
☆32Updated 4 months ago
Alternatives and similar repositories for Tiled-MM
Users that are interested in Tiled-MM are comparing it to the libraries listed below
Sorting:
- ☆45Updated 4 years ago
- rocSHMEM intra-kernel networking runtime for AMD dGPUs on the ROCm platform.☆97Updated this week
- Slides and exercises for persistent memory programming tutorial☆14Updated 2 years ago
- High Performance Linpack for Next-Generation AMD HPC Accelerators☆60Updated 2 weeks ago
- An HPL-AI implementation for Fugaku☆21Updated 4 years ago
- Advanced Profiling and Analytics for AMD Hardware☆161Updated this week
- Prototype of OpenSHMEM for NVIDIA GPUs, developed as part of DoE Design Forward☆25Updated 7 years ago
- ☆18Updated 3 years ago
- A task benchmark☆43Updated 11 months ago
- A Micro-benchmarking Tool for HPC Networks☆32Updated last week
- 🎃 GPU load-balancing library for regular and irregular computations.☆62Updated last year
- A hierarchical collective communications library with portable optimizations☆36Updated 7 months ago
- ☆18Updated last year
- Benchmark for measuring the performance of sparse and irregular memory access.☆78Updated 2 months ago
- The Task-Aware MPI (TAMPI) library extends the functionality of standard MPI libraries by providing new mechanisms for improving the inte…☆24Updated last month
- A memory profiler for NVIDIA GPUs to explore memory inefficiencies in GPU-accelerated applications.☆25Updated 9 months ago
- RCCL Performance Benchmark Tests☆71Updated this week
- GPU Performance Advisor☆65Updated 3 years ago
- Using C++ magic to launch/capture CUDA kernels and tune them with Kernel Tuner☆20Updated last year
- An extension library of WMMA API (Tensor Core API)☆99Updated last year
- ROC profiler library. Profiling with perf-counters and derived metrics.☆151Updated 2 weeks ago
- Graph-indexed Pandas DataFrames for analyzing hierarchical performance data☆34Updated 3 weeks ago
- NUMA-aware multi-CPU multi-GPU data transfer benchmarks☆24Updated last year
- A tracing infrastructure for heterogeneous computing applications.☆34Updated last week
- TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)☆44Updated this week
- NAS Parallel Benchmarks for evaluating GPU and APIs☆26Updated last week
- AMD’s C++ library for accelerating tensor primitives☆43Updated last week
- Very-Low Overhead Checkpointing System☆58Updated this week
- Bandwidth test for ROCm☆62Updated this week
- Reference implementations of MLPerf™ HPC training benchmarks☆48Updated 5 months ago