Nikolay

I experience the same problem with: - SynapseML Version: v0.9.5 - Spark Version 3.2.0 - Spark Platform on-premise k8s The problem is not regular and happens from time to time.

Hi @Ishitori, is there a version of synapseml 0.9.5 that supports Spark 3.0.1 (in particular, with working saving of trained LIghtGBM)? Or has its support been completely dropped already?

@imatiach-msft @svotaw if I understand correctly, using native memory isn't counted by JVM at all and it may lead to OOM errors when running on some resource managers (Kubernetes, for...