Error-bounded lossy compression has been effective in significantly reducing the data
storage/transfer burden while preserving the reconstructed data fidelity very well. Many error …

Existing GPU lossy compressors suffer from expensive data movement overheads,
inefficient memory access patterns, and high synchronization latency, resulting in limited …

Error-bounded lossy compression has been a critical technique to significantly reduce the
sheer amounts of simulation datasets for high-performance computing (HPC) scientific …

As network bandwidth struggles to keep up with rapidly growing computing capabilities, the
efficiency of collective communication has become a critical challenge for exa-scale …

Recent years have witnessed a clear trend towards language models with an ever-
increasing number of parameters, as well as the growing training overhead and memory …

In the era of exascale computing, the adoption of a large number of CPU cores and nodes
by high-performance computing (HPC) applications has made MPI collective performance …

Abstract The NVIDIA Collective Communications Library (NCCL) is a multi-GPU
communication library widely used in applications such as deep learning, molecular …

Data acquisition (DAQ) networks, widely used in scientific research and industrial
applications, are composed of numerous interconnected servers, exchanging substantial …