Cite
Notes
Only stored in your browser.
Attribution
Harmony: Overcoming the Hurdles of GPU Memory Capacity to Train Massive DNN Models on Commodity Servers
arXiv 2022
PipeDream: Fast and Efficient Pipeline Parallel DNN Training
arXiv 2018
from 2 papers
Aaron Harlap
Deepak Narayanan
Derek Murray
Greg Ganger
Jakub Tarnawski
Nam Sung Kim
Nikhil Devanur
Phil Gibbons
Vivek Seshadri
Youjie Li