Sign in

SWIFT: Expedited Failure Recovery for Large-scale DNN Training.

Yuchen ZhongGuangming ShengJuncheng LiuJinhui YuanChuan Wu
Published in: CoRR (2023)
Keyphrases
  • failure recovery
  • training process
  • load balancing
  • fault tolerance
  • single link
  • management system
  • multimedia
  • query optimization
  • grid computing