Due to the high overheating warning on GPU clusters, some parts of cluster will needed to be exchanged. We are currently working on it with other teams.
Meanwhile, it is advised to avoid any running jobs on GPU001 and GPU003. Please kill all the running jobs on two nodes, if any. And please DON’T run any process on GPU001 and GPU003 until further notification. We will let you know with any updates.
Thank you for your patience!
SOLVED by August 8th, 2024
– RebelX Team