You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@spark.apache.org by Fatma Ozcan <fa...@gmail.com> on 2015/09/19 03:37:10 UTC

SparkML pipelines and error recovery

Trying to understand how Spark ML pipelines work in case of failures. If I
have multiple transformers and one of them fails, will the lineage based
recovery of rdd's automatically kick in?

Thanks,
Fatma