Caesar AI Atlas

Pipelining

Caesar AI Atlas Definition

Pipelining is a form of model parallelism that splits model processing into sequential stages across different devices. It improves throughput by allowing different stages to process different batches at the same time.

Other Definitions

Pipelining Source

A form of model parallelism in which a model's processing is divided into consecutive stages and each stage is executed on a different device. While a stage is processing one batch, the preceding stage can work on the next batch. See also staged training.

Related Terms