2026.03.21[Restored] Regarding the Failure of the Interdisciplinary Large-Scale Computing System
At approximately 3:50 a.m. on March 21, 2026, an instantaneous voltage dip—presumed to have been caused by strong winds—occurred, resulting in electrical system irregularities that have affected portions of the computing system and the cooling infrastructure.
We are currently conducting a detailed assessment of the situation and implementing necessary countermeasures; however, at this time, we are unable to provide an estimate for full restoration.
We will provide further updates as soon as there is any significant progress.
(Added on March 21, 2026, 19:00)
Due to this incident, a number of compute nodes within the supercomputing system (both CPU and GPU nodes) were brought offline.
The cooling system has already been fully restored, and we are currently proceeding with the recovery of the affected nodes as well as verifying the extent of the impact.
Once restoration is complete and the scope of the impact has been determined, the Center will contact individual users whose running jobs were affected.
Please note that the impact of this incident is limited strictly to the supercomputing system; the Research cloud and Nextcloud services have not been affected.
Login access to the supercomputer and the submission of new jobs remain available; however, jobs will begin executing sequentially on nodes as they are restored.
We sincerely apologize for the inconvenience this incident has caused to all users.
We kindly ask for your patience and understanding as we work toward complete recovery.
Service was restored at 12:45 a.m. on March 24.