Messaging and state layer for distributed serverless applications
-
Updated
May 13, 2024 - C++
Messaging and state layer for distributed serverless applications
Distributed training (multi-node) of a Transformer model
Blink+: Increase GPU group bandwidth by utilizing across tenant NVLink.
collectives library for upc++
A Distributed Parallel Training Simulation Tool (AdpartSim) for Data Center focuses on helping us study and simulate the parallel optimization strategies of Large Models (LM), as well as the impact of network topology and collective communication on the training efficiency of LM.
A reduction algorithm for MPI using only peer to peer communication
Modelling of MPI collective operations latencies: Broadcast and Reduce operations. UniTS, SDIC, 2023-2024
This repository contains simple programs of MPI_Bcast, MPI_Reduce, MPI_Scatter and MPI_Gather. Download the repository and test your self.
Add a description, image, and links to the collective-communication topic page so that developers can more easily learn about it.
To associate your repository with the collective-communication topic, visit your repo's landing page and select "manage topics."