TrainMover achieves ~20s downtime for interruptions in 1024-GPU LLM training via two-phase delta-based communication setup, communication-free sandboxed warmup, and general standby design, projecting 55% reduction in wasted GPU hours.
Automated tensor model paral- lelism with overlapped communication for efficient foundation model training, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
TrainMover: An Interruption-Resilient Runtime for ML Training
TrainMover achieves ~20s downtime for interruptions in 1024-GPU LLM training via two-phase delta-based communication setup, communication-free sandboxed warmup, and general standby design, projecting 55% reduction in wasted GPU hours.