Glossary

Pipeline bubble

Beginner

Waiting time on an assembly line: at the start, later stations have nothing to do yet, and at the end, early ones have finished.

Novice

An empty slot moving through a pipeline, where a stage has no useful work. In a CPU it comes from a stall or a flushed instruction; in pipeline-parallel training it is the idle time while the pipeline fills at the start of a step and drains at the end, which shrinks when there are many more microbatches than stages.

Expert

In a CPU pipeline, a no-op inserted by a stall or a squash, costing one cycle of CPI per bubble. In pipeline-parallel training, (p−1)(tf+tb)(p-1)(t_{\mathrm{f}} + t_{\mathrm{b}}) of idle time per step with flushes, a fraction (p−1)/m(p-1)/m of ideal compute time; interleaving vv model chunks per device divides it by vv at vv times the point-to-point traffic.