Who is down right now

Nebius · AIMay 13, 2026, 18:45 UTC 5 months ago

InfiniBand performance degradation

MinorNot disclosedUpdated 2h ago
May 13, 18:45 UTCMay 13, 21:05 UTC
Duration
2h 20m
Impact
Minor
Root cause
Not disclosed
Nebius, 90 days
22 incidents
Affected
Compute CloudUK-SOUTH1 - Compute Cloud
Status page

Final update

This incident has been resolved.

Timeline

  1. Resolved · May 13, 21:05 UTC
    This incident has been resolved.
  2. Investigating · May 13, 18:45 UTC
    We are investigating reduced InfiniBand fabric performance affecting GPU workloads in the uk-south1 region. Customers may observe lower throughput and slower distributed training jobs.

More from Nebius

Full history
MinorOct 8, 08:48 UTC
Metrics unavailable in Management Console for Monitoring in eu-west1lasted 11m
MajorOct 7, 16:37 UTC
Problems with power for several dozens of nodeslasted 14h 45m
MinorOct 4, 15:54 UTC
Issues with preempted Managed Kubernetes nodes in all regionslasted 24h
MinorOct 3, 02:30 UTC
InfiniBand connectivity issues for GPU clusters in eu-west2lasted 6h 28m
MajorOct 2, 01:58 UTC
Compute VM creation and Object storage are partial unavailablelasted 9m
MajorSep 23, 05:20 UTC
Network issues in eu-west2lasted 3h 12m