Donglin Yang

Publications

Redundancy-Free High-Performance Dynamic GNN Training with Hierarchical Pipeline Parallelism.

ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC), 2023.

MPipeMoE: Memory Efficient MoE for Pre-trained Models with Adaptive Pipeline Parallelism.

The 37th IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2023.

Mitigating Stragglers in the Decentralized Training on Heterogeneous Clusters

Donglin Yang, Dazhao Cheng, Wei Rang.

ACM/IFIP International Conference on Middleware (Middleware), 2020.

Efficient GPU Memory Management for Nonlinear DNNs.

Donglin Yang, Dazhao Cheng.

ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC), 2020.

Elastic Executor Provisioning for Iterative Workloads on Apache Spark

Donglin Yang, Wei Rang, Dazhao Cheng, Yu Wang, Jiannan Tian, and Dingwen Tao.

IEEE International Conference on Big Data (BigData), 2019.

Joint Optimization of MapReduce Scheduling and Network Policy in Hierarchical Clouds

Donglin Yang, Wei Rang, Dazhao Cheng.

ACM International Conference on Parallel Processing (ICPP), 2018.

Joint Optimization of MapReduce Scheduling and Network Policy in Hierarchical Data Centers

Donglin Yang, Wei Rang, Dazhao Cheng, Yu Wang.

IEEE Transactions on Cloud Computing.

Scalable Data Management on Hybrid Memory System for Deep Neural Network Applications

Wei Rang, Donglin Yang, Zhimin Li, Dazhao Cheng.

IEEE International Conference on Big Data, 2021.

Data Life Aware Model Updating Strategy for Stream-based Online Deep Learning

Wei Rang, Donglin Yang, Dazhao Cheng, Kun Suo, Wei Chen.

IEEE Transactions on Parallel and Distributed Systems.

Dependency-aware Tensor Scheduler for Industrial AI Applications

Wei Rang, Donglin Yang, Dazhao Cheng.

IEEE Industrial Electronics Magazine.

A Shared Memory Cache Layer across Multiple Executors in Apache Spark

Wei Rang, Donglin Yang, Dazhao Cheng.

IEEE International Conference on Big Data, 2020.