CheckFreq: Frequent, Fine-Grained DNN Checkpointing.

Jayashree Mohan Amar Phanishayee Vijay Chidambaram

Published in: FAST (2021)

Keyphrases

fine grained
coarse grained
distributed databases
distributed database systems
main memory databases
fault tolerance
concurrency control
access control
tightly coupled
low overhead
data lineage
data provenance
database systems
training process
frequent patterns
data structure
web services
failure recovery
database