A new open-source library called Alpa aims to automate distributed training and serving of large deep networks. It proposes a compiler where existing model-parallel strategies are combined and the usa ...

This was a high-stakes implementation of advanced data pipelines designed with zero tolerance for data inconsistency or ...

The Parallel & Distributed Computing Lab (PDCL) conducts research at the intersection of high performance computing and big data processing. Our group works in the broad area of Parallel & Distributed ...