lsds / KungFu
Fast and Adaptive Distributed Machine Learning for TensorFlow, PyTorch and MindSpore.
☆293Updated last year
Alternatives and similar repositories for KungFu:
Users that are interested in KungFu are comparing it to the libraries listed below
- Simple Distributed Deep Learning on TensorFlow☆134Updated 2 years ago
- CS294; AI For Systems and Systems For AI☆225Updated 5 years ago
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆127Updated 2 years ago
- Fine-grained GPU sharing primitives☆141Updated 5 years ago
- ☆389Updated 2 years ago
- GPU-scheduler-for-deep-learning☆203Updated 4 years ago
- Kubernetes Scheduler for Deep Learning☆260Updated 2 years ago
- ☆83Updated 2 years ago
- Model-less Inference Serving☆85Updated last year
- Easy Parallel Library (EPL) is a general and efficient deep learning framework for distributed model training.☆267Updated last year
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆74Updated 4 years ago
- A lightweight parameter server interface☆75Updated 2 years ago
- ☆580Updated 6 years ago
- Code for "Heterogenity-Aware Cluster Scheduling Policies for Deep Learning Workloads", which appeared at OSDI 2020☆126Updated 7 months ago
- Resource-adaptive cluster scheduler for deep learning training.☆436Updated 2 years ago
- Tiresias is a GPU cluster manager for distributed deep learning training.☆152Updated 4 years ago
- MLModelCI is a complete MLOps platform for managing, converting, profiling, and deploying MLaaS (Machine Learning-as-a-Service), bridging…☆192Updated 2 years ago
- ☆186Updated 5 years ago
- A tensor-aware point-to-point communication primitive for machine learning☆256Updated 2 years ago
- Serverless ML Framework☆107Updated 2 years ago
- A high-performance framework for training wide-and-deep recommender systems on heterogeneous cluster☆157Updated 11 months ago
- Elastic Deep Learning for deep learning framework on Kubernetes☆172Updated last year
- ☆214Updated last year
- ☆51Updated last year
- 🔮 Execution time predictions for deep neural network training iterations across different GPUs.☆60Updated 2 years ago
- A Tool for Automatic Parallelization of Deep Learning Training in Distributed Multi-GPU Environments.☆130Updated 3 years ago
- Fault-tolerant for DL frameworks☆69Updated last year
- A Deep Learning Cluster Scheduler☆37Updated 4 years ago
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆116Updated last year
- ☆372Updated 7 years ago