feifeibear / PSTensor
PSTensor provides a way to hack the memory management of tensors in TensorFlow and PyTorch by defining your own C++ Tensor Class.
☆10Updated 3 years ago
Alternatives and similar repositories for PSTensor:
Users that are interested in PSTensor are comparing it to the libraries listed below
- An external memory allocator example for PyTorch.☆14Updated 3 years ago
- ☆44Updated last year
- ☆8Updated last year
- Inference framework for MoE layers based on TensorRT with Python binding☆41Updated 3 years ago
- Fairring (FAIR + Herring) is a plug-in for PyTorch that provides a process group for distributed training that outperforms NCCL at large …☆63Updated 2 years ago
- Tacker: Tensor-CUDA Core Kernel Fusion for Improving the GPU Utilization while Ensuring QoS☆19Updated last month
- [ICDCS 2023] DeAR: Accelerating Distributed Deep Learning with Fine-Grained All-Reduce Pipelining☆12Updated last year
- High performance NCCL plugin for Bagua.☆15Updated 3 years ago
- An Attention Superoptimizer☆21Updated last month
- ☆72Updated 3 years ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆38Updated 2 years ago
- ☆19Updated 5 months ago
- Official resporitory for "IPDPS' 24 QSync: Quantization-Minimized Synchronous Distributed Training Across Hybrid Devices".☆19Updated last year
- PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections☆119Updated 2 years ago
- Thunder Research Group's Collective Communication Library☆33Updated 10 months ago
- Dynamic Tensor Rematerialization prototype (modified PyTorch) and simulator. Paper: https://arxiv.org/abs/2006.09616☆133Updated last year
- ☆11Updated 3 years ago
- An extention of TVMScript to write simple and high performance GPU kernels with tensorcore.☆51Updated 7 months ago
- ☆24Updated 2 months ago
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆53Updated 6 months ago
- ☆84Updated 6 months ago
- 🔮 Execution time predictions for deep neural network training iterations across different GPUs.☆60Updated 2 years ago
- Slides from 2021-12-15 talk, "TVM Developer Bootcamp – Writing Hardware Backends"☆11Updated 3 years ago
- Code for Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture (accepted by PVLDB).The outdated wr…☆9Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing. By pro…☆67Updated this week
- FTPipe and related pipeline model parallelism research.☆41Updated last year
- Odysseus: Playground of LLM Sequence Parallelism☆66Updated 8 months ago
- Cavs: An Efficient Runtime System for Dynamic Neural Networks☆14Updated 4 years ago
- ☆26Updated 3 years ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆25Updated this week