Tensorflow Straggler Mitigation by Active Backup Worker

ShifuML / shifu

An end-to-end machine learning and data mining framework on Hadoop

https://github.com/ShifuML/shifu/wiki

Apache License 2.0

251 stars 109 forks source link

Tensorflow Straggler Mitigation by Active Backup Worker #643

Open zhangpengshan opened 5 years ago

zhangpengshan commented 5 years ago

Current backup worker is only running backup script in wait state like backup.py, while it is better to run active backup to random get split data to run and contribute to each iteration.